{
    "ok": true,
    "version": "3.0.0-mission-control",
    "checkedAt": "2026-07-23T01:58:07+00:00",
    "elapsedMs": 237,
    "count": 20,
    "categories": {
        "ai": 20
    },
    "items": [
        {
            "id": "70291f8c9155d1e1120ce6504e8dbd5227209441",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "SysAdmin: Measuring Instrumental Power-Seeking in Frontier AI",
            "summary": "arXiv:2607.18239v1 Announce Type: new Abstract: Power-seeking defined as behaviors where AI systems acquire resources, evade oversight, or resist termination beyond task requirements is identified as a key driver of Loss of Control (LoC) risk. In this work, we introduce SysAdmin, a benchmark that positions frontier language models as autonomous system administrators in a high-fidelity Linux sandbox to measure power-seeking propensity across five dimensions: self-preservation, increasing autonomy, resource acquisition, environment modification, and strategic concealment. We evaluated seven frontier models across four experimental conditions in a total of 2800 tasks. After bias correction using human-annotated calibration data, corrected power-seeking estimates ranged from 0 to about 5 percent per model. We also conducted a positive control with explicit power-seeking prompts that achieve…",
            "url": "https://arxiv.org/abs/2607.18239",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "SysAdmin: Measuring Instrumental Power-Seeking in Frontier AI",
                "summary": "arXiv:2607.18239v1 Announce Type: new \nAbstract: Power-seeking defined as behaviors where AI systems acquire resources, evade oversight, or resist termination beyond task requirements is identified as a key driver of Loss of Control (LoC) risk. In this work, we introduce SysAdmin, a benchmark that positions frontier language models as autonomous system administrators in a high-fidelity Linux sandbox to measure power-seeking propensity across five dimensions: self-preservation, increasing autonomy, resource acquisition, environment modification, and strategic concealment. We evaluated seven frontier models across four experimental conditions in a total of 2800 tasks. After bias correction using human-annotated calibration data, corrected power-seeking estimates ranged from 0 to about 5 percent per model. We also conducted a positive control with explicit power-seeking prompts that achieved 100% detection, validating measurement sensitivity. Our findings indicate current frontier models exhibit minimal spontaneous power-seeking in naturalistic system administration contexts, though model-specific failure modes suggest evaluations must test diverse misalignment patterns. Nevertheless, we discovered other more pronounced failure modes (than power-seeking) such as specification gaming and resistance to goal modification.",
                "url": "https://arxiv.org/abs/2607.18239",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "c2f64f638337cfb27cc234f2b988a1af19718adf",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Calibrated Selective Fact-Checking via Evidence Chain Evaluation",
            "summary": "arXiv:2607.18240v1 Announce Type: new Abstract: Large language models (LLMs) can achieve strong fact-checking accuracy, yet forced binary decisions conceal a critical reliability problem: systems may issue confident verdicts even when supporting evidence is weak, sparse, or internally inconsistent. We address this issue through Evidence Chain Evaluation (ECE), a selective fact-checking framework that permits abstention via an uncertain verdict instead of requiring a true/false decision for every claim. The evaluated system is a tool-using verification agent that gathers evidence through web search, scholarly search, and executable checks, and then returns a structured verdict with confidence and source-level metadata. On ECE-Bench, ECE achieves 91.6% standard accuracy, 93.7% coverage, and 97.8% selective accuracy on answered claims. Although ECE does not outperform the strongest retriev…",
            "url": "https://arxiv.org/abs/2607.18240",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Calibrated Selective Fact-Checking via Evidence Chain Evaluation",
                "summary": "arXiv:2607.18240v1 Announce Type: new \nAbstract: Large language models (LLMs) can achieve strong fact-checking accuracy, yet forced binary decisions conceal a critical reliability problem: systems may issue confident verdicts even when supporting evidence is weak, sparse, or internally inconsistent. We address this issue through Evidence Chain Evaluation (ECE), a selective fact-checking framework that permits abstention via an uncertain verdict instead of requiring a true/false decision for every claim. The evaluated system is a tool-using verification agent that gathers evidence through web search, scholarly search, and executable checks, and then returns a structured verdict with confidence and source-level metadata. On ECE-Bench, ECE achieves 91.6% standard accuracy, 93.7% coverage, and 97.8% selective accuracy on answered claims. Although ECE does not outperform the strongest retrieval baseline on aggregate calibration metrics such as Expected Calibration Error, Brier score, or AURC, it delivers a clear selective-prediction trade-off: the system maintains very high accuracy on answered claims while deferring 6 of 95 cases. These deferred cases are concentrated in lower-reliability evidence settings (5/6 at source level L4), supporting the view that abstention functions as a safety-oriented mechanism for handling epistemically weak evidence. Code is available at https://github.com/ cheshireyang/ECE.git",
                "url": "https://arxiv.org/abs/2607.18240",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "9f4fe4236892165000693cf3559f6f019574be8a",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "BatchDAG: LLM-Planned Execution Graphs for Scalable Ad-Hoc Analysis Over Enterprise Data",
            "summary": "arXiv:2607.18241v1 Announce Type: new Abstract: Large language models (LLMs) excel at analyzing individual documents but break down on exhaustive, cross-entity analytical questions over enterprise-scale datasets due to context overflow, loss of per-entity attribution, and linear latency from sequential tool calls. We present BatchDAG, a system in which an LLM generates a typed directed acyclic graph (DAG) of operations -- SQL queries, semantic searches, in-memory transforms, parallel fan-outs, and single-shot analyses -- which a deterministic engine evaluates with topological-wave parallelism and structured JSON data flow. A key optimization, entity-aware batching, groups rows by logical entity before fan-out, reducing LLM calls by up to 47x. BatchDAG is not primarily an accuracy improvement over hand-optimized pipelines; rather, it is a general-purpose orchestration layer that replaces…",
            "url": "https://arxiv.org/abs/2607.18241",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "BatchDAG: LLM-Planned Execution Graphs for Scalable Ad-Hoc Analysis Over Enterprise Data",
                "summary": "arXiv:2607.18241v1 Announce Type: new \nAbstract: Large language models (LLMs) excel at analyzing individual documents but break down on exhaustive, cross-entity analytical questions over enterprise-scale datasets due to context overflow, loss of per-entity attribution, and linear latency from sequential tool calls. We present BatchDAG, a system in which an LLM generates a typed directed acyclic graph (DAG) of operations -- SQL queries, semantic searches, in-memory transforms, parallel fan-outs, and single-shot analyses -- which a deterministic engine evaluates with topological-wave parallelism and structured JSON data flow. A key optimization, entity-aware batching, groups rows by logical entity before fan-out, reducing LLM calls by up to 47x. BatchDAG is not primarily an accuracy improvement over hand-optimized pipelines; rather, it is a general-purpose orchestration layer that replaces multiple hand-engineered workflows with a single system that generates the appropriate execution strategy from natural language. In controlled experiments on 12 transcript-heavy queries, BatchDAG (3.74/5) achieves quality comparable to an expert-designed pipeline (3.25/5) and significantly outperforms a ReAct agent (3.09/5, p<0.01), with superior provenance (77% transcript evidence rate vs. 46-60% for baselines). A controlled ablation shows structured JSON intermediates reduce hallucinations by 27% versus prose summaries (paired t-test, p=0.107, n=12). The planner achieves 98.8% valid-DAG rate across 300 planning calls. In production at Brevian.ai, BatchDAG processes queries over 50,000+ meetings in under 60 seconds, with measured per-query costs of $0.02-$0.24 at published GPT-5.1 pricing.",
                "url": "https://arxiv.org/abs/2607.18241",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "7ad17dfb19d052217ffd1a4e914433e5ac541f7f",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "AI Tool Discovery at Scale: All You Need is DNS",
            "summary": "arXiv:2607.18242v1 Announce Type: new Abstract: The coming era of autonomous AI agents demands a discovery mechanism capable of navigating millions of tools, yet existing solutions buckle under O(N) complexity and centralized governance. Instead of building another fragile overlay, we propose ToolDNS, a radical framework that retrofits semantic tool discovery onto the Internet's most resilient substrate: the Domain Name System (DNS). By embedding functional intent and organizational trust into a hierarchical namespace, ToolDNS transforms an expensive semantic search into a series of lightweight, O(log N) name resolutions. We introduce three protocol-compliant enhancements to enable decentralized governance and semantic pruning: partially unfolded names, EDNS0 intent payloads, and logical subdomains. To rigorously evaluate this approach across the fragmented tooling landscape, we constru…",
            "url": "https://arxiv.org/abs/2607.18242",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "AI Tool Discovery at Scale: All You Need is DNS",
                "summary": "arXiv:2607.18242v1 Announce Type: new \nAbstract: The coming era of autonomous AI agents demands a discovery mechanism capable of navigating millions of tools, yet existing solutions buckle under O(N) complexity and centralized governance. Instead of building another fragile overlay, we propose ToolDNS, a radical framework that retrofits semantic tool discovery onto the Internet's most resilient substrate: the Domain Name System (DNS). By embedding functional intent and organizational trust into a hierarchical namespace, ToolDNS transforms an expensive semantic search into a series of lightweight, O(log N) name resolutions. We introduce three protocol-compliant enhancements to enable decentralized governance and semantic pruning: partially unfolded names, EDNS0 intent payloads, and logical subdomains. To rigorously evaluate this approach across the fragmented tooling landscape, we construct and release a large-scale heterogeneous benchmark comprising 33,688 real-world tools spanning MCP, A2A, RESTful, and Skill protocols. On this dataset, ToolDNS slashes the per-query search space by 95.26% while matching state-of-the-art retrieval accuracy. Furthermore, its UDP-native design reduces discovery latency by orders of magnitude compared to HTTP-based registries. Our work demonstrates that scalable AI interoperability requires not more middleware, but a smarter utilization of the infrastructure already beneath our feet.",
                "url": "https://arxiv.org/abs/2607.18242",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "b7fa891ce5fc80ac73fa5f29974d875fdff62c7a",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "From Agent Failure Paths to Quantified Residual Risk: A Compositional Framework for Resilient Agentic AI",
            "summary": "arXiv:2607.18243v1 Announce Type: new Abstract: Agentic AI is crossing trust boundaries faster than current risk models can represent. Existing approaches provide one of two partial views. They either describe failure mechanisms without producing a transferable residual-risk estimate, or they produce a risk estimate while treating the internal failure path as a black box. We couple those two views by proposing CPSAINT, a seven-layer integrity decomposition over Physical state, Sensors, Data, Compute, Actuators, Environment, and Time, paired with FRIESA-K, a residual-risk functional that maps each failure path to a quantified risk instance. FRIESA-K grounds the resistance term K in a controlled absorbing Markov model so that control effectiveness is derived from state dynamics rather than assigned as an informal score. The result is a concise mechanism-to magnitude pipeline for resilient…",
            "url": "https://arxiv.org/abs/2607.18243",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "From Agent Failure Paths to Quantified Residual Risk: A Compositional Framework for Resilient Agentic AI",
                "summary": "arXiv:2607.18243v1 Announce Type: new \nAbstract: Agentic AI is crossing trust boundaries faster than current risk models can represent. Existing approaches provide one of two partial views. They either describe failure mechanisms without producing a transferable residual-risk estimate, or they produce a risk estimate while treating the internal failure path as a black box. We couple those two views by proposing CPSAINT, a seven-layer integrity decomposition over Physical state, Sensors, Data, Compute, Actuators, Environment, and Time, paired with FRIESA-K, a residual-risk functional that maps each failure path to a quantified risk instance. FRIESA-K grounds the resistance term K in a controlled absorbing Markov model so that control effectiveness is derived from state dynamics rather than assigned as an informal score. The result is a concise mechanism-to magnitude pipeline for resilient agentic and embodied AI. We report governance observability through a separate additive penalty instead of inserting governance as a new variable in the resistance functional. We formalize structural composability linking valid failure paths to well-defined risk instances and show the framework on two contrasting scenarios a hard real-time warehouse robot and a governance-instrumented financial-services agent. Across both cases, the same layer grammar, variable semantics, and dynamic-resistance construction remain intact. Thus, we obtain a compact kernel that supports cross-domain reasoning, explicit assumptions, and quantitatively grounded formalism of composable trust.",
                "url": "https://arxiv.org/abs/2607.18243",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "723ba34a07a579f4abe0481932fc811f0b966765",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "SAAG: Structured Agent Assessment and Grounding",
            "summary": "arXiv:2607.18245v1 Announce Type: new Abstract: Exact-match evaluation of agent-calling obscures qualitatively different failure modes: a model may select the right function yet hallucinate argument values, or satisfy a schema while choosing a agent for the wrong reason. Existing benchmarks collapse these distinctions into a single binary score, leaving practitioners unable to diagnose where agent calls fail. We propose SAAG a cascaded diagnostic framework that decomposes agent-calling evaluation into three sequential stages: registry conformance, structural completeness, and argument grounding, each producing interpretable stage-specific diagnostics. These diagnostics additionally enable iterative self-repair: on prediction failure, the stage-specific signal guides targeted correction without leaking ground-truth values. We evaluate this framework on a controlled benchmark derived from…",
            "url": "https://arxiv.org/abs/2607.18245",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "SAAG: Structured Agent Assessment and Grounding",
                "summary": "arXiv:2607.18245v1 Announce Type: new \nAbstract: Exact-match evaluation of agent-calling obscures qualitatively different failure modes: a model may select the right function yet hallucinate argument values, or satisfy a schema while choosing a agent for the wrong reason. Existing benchmarks collapse these distinctions into a single binary score, leaving practitioners unable to diagnose where agent calls fail. We propose SAAG a cascaded diagnostic framework that decomposes agent-calling evaluation into three sequential stages: registry conformance, structural completeness, and argument grounding, each producing interpretable stage-specific diagnostics. These diagnostics additionally enable iterative self-repair: on prediction failure, the stage-specific signal guides targeted correction without leaking ground-truth values. We evaluate this framework on a controlled benchmark derived from Glaive's function-calling dataset across registry sizes of 5, 10, and 15 agents using three local sub-4B-parameter models. Structured feedback consistently improves argument precision and reduces value hallucination relative to single-pass inference and uninformative binary feedback, while end-to-end F1 gains are modest and model-dependent. These results suggest that stage-decomposed diagnostic evaluation is a necessary lens for understanding and improving agent-calling reliability across model families and registry scales.",
                "url": "https://arxiv.org/abs/2607.18245",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "5be44a3436a044c46c7553301e39f2dab7edf184",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Phionyx: A Deterministic AI Runtime Architecture with Structured State Management and Pre-Response Governance",
            "summary": "arXiv:2607.18246v1 Announce Type: new Abstract: We present Phionyx, a deterministic AI runtime architecture derived from the broader Echoism interaction framework that introduces a governance-first approach to AI engineering: treating large language model (LLM) outputs as noisy sensor measurements rather than direct decisions. Unlike probabilistic agents, Phionyx enforces deterministic state evolution via a structured state vector governed by deterministic state-evolution equations, enabling reproducible behavior in applications requiring auditability and governance. The architecture integrates three layers: (1) a deterministic evaluation kernel processing noisy sensor measurements through a canonical 46-block pipeline, (2) a unified safety layer providing pre-response control and architectural privacy enforcement, and (3) a semantic time-based memory system implementing impact-weighted…",
            "url": "https://arxiv.org/abs/2607.18246",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Phionyx: A Deterministic AI Runtime Architecture with Structured State Management and Pre-Response Governance",
                "summary": "arXiv:2607.18246v1 Announce Type: new \nAbstract: We present Phionyx, a deterministic AI runtime architecture derived from the broader Echoism interaction framework that introduces a governance-first approach to AI engineering: treating large language model (LLM) outputs as noisy sensor measurements rather than direct decisions. Unlike probabilistic agents, Phionyx enforces deterministic state evolution via a structured state vector governed by deterministic state-evolution equations, enabling reproducible behavior in applications requiring auditability and governance. The architecture integrates three layers: (1) a deterministic evaluation kernel processing noisy sensor measurements through a canonical 46-block pipeline, (2) a unified safety layer providing pre-response control and architectural privacy enforcement, and (3) a semantic time-based memory system implementing impact-weighted cache eviction. Experimental validation on single-instance deployments demonstrates approximately 31% reduction in computational overhead vs. post-hoc filtering (at 30% unsafe input ratio, simulated cost model) and up to 24% improvement in high-value data retention vs. LRU (72% vs. FIFO, same cache capacity, benchmark-verified), deterministic execution verified across 100 repeated runs with zero variance in control signals (hash-verified), and zero unplanned restarts in single-instance deployment testing (see Appendix C for methodology and scope). This paper presents the architecture, its analytic structure, and scoped experimental evidence; generalization to distributed or multi-tenant deployments remains future work.",
                "url": "https://arxiv.org/abs/2607.18246",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "b85d9aff9f6fbe32925fa0f81b18c31688cfec0b",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Integro-differential equations in angular stabilization of drone motion by distributed feedback control",
            "summary": "arXiv:2607.18251v1 Announce Type: new Abstract: In this paper, we propose angular stabilization of drone motion using distributed feedback control in the form of an integral operator. It should be stressed that the memory of this integral operator could be unbounded. It is intuitively clear that large length of the observation time open new possibilities to construct better control based on previous states of the control object. Unbounded memory in control requires the creation of a certain approach different from standard ones to the study of integro-differential equations. One of the goals of this article is to propose a certain universal approach that allows us to study the stability of integro-differential equations in the case of unbounded memory in the integral operator specifying the feedback control in stabilization. The approach we propose allows us to reduce the study of integ…",
            "url": "https://arxiv.org/abs/2607.18251",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Integro-differential equations in angular stabilization of drone motion by distributed feedback control",
                "summary": "arXiv:2607.18251v1 Announce Type: new \nAbstract: In this paper, we propose angular stabilization of drone motion using distributed feedback control in the form of an integral operator. It should be stressed that the memory of this integral operator could be unbounded. It is intuitively clear that large length of the observation time open new possibilities to construct better control based on previous states of the control object. Unbounded memory in control requires the creation of a certain approach different from standard ones to the study of integro-differential equations. One of the goals of this article is to propose a certain universal approach that allows us to study the stability of integro-differential equations in the case of unbounded memory in the integral operator specifying the feedback control in stabilization. The approach we propose allows us to reduce the study of integro-differential equations to the analysis of systems of ordinary differential equations. In general, such systems can consist of an infinite number of equations. In relation to the so-called linear approximation in the problem of angle stabilization manages to limit itself to relatively simple exponential kernels in the integral control and arrive at a system with a finite number of equations. The examples explain that more complex kernels, for example, linear combinations of the exponential kernels, can enhance the stabilization capabilities. We obtain new unexpectable results on the exponential stability of integro-differential equations. Then we apply them to stabilization of drone flight.",
                "url": "https://arxiv.org/abs/2607.18251",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "52fda716aaa5413f55642e7465aeb96f9554908d",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "MILP-Evo: Closed-Loop Fully Automatic Design of MILP Solvers",
            "summary": "arXiv:2607.18252v1 Announce Type: new Abstract: Machine learning methods have shown that data-driven policies can accelerate mixed-integer linear programming (MILP) solvers, but many such approaches remain difficult to inspect, adapt, and deploy because the learned policy is represented as an external predictor or other opaque model. By contrast, explicit solver logic is easier to understand and integrate, but is usually hand-designed rather than learned from solver feedback. We study whether the automatic design of MILP solver logic can instead be cast as LLM-guided closed-loop search over executable white-box components evaluated directly by end-to-end solver behavior. To this end, we propose a closed-loop program evolution framework for MILP solver auto-design, implemented through PySCIPOpt, and instantiate it on the joint design of a cut selector and a branching rule. Candidate prog…",
            "url": "https://arxiv.org/abs/2607.18252",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "MILP-Evo: Closed-Loop Fully Automatic Design of MILP Solvers",
                "summary": "arXiv:2607.18252v1 Announce Type: new \nAbstract: Machine learning methods have shown that data-driven policies can accelerate mixed-integer linear programming (MILP) solvers, but many such approaches remain difficult to inspect, adapt, and deploy because the learned policy is represented as an external predictor or other opaque model. By contrast, explicit solver logic is easier to understand and integrate, but is usually hand-designed rather than learned from solver feedback. We study whether the automatic design of MILP solver logic can instead be cast as LLM-guided closed-loop search over executable white-box components evaluated directly by end-to-end solver behavior. To this end, we propose a closed-loop program evolution framework for MILP solver auto-design, implemented through PySCIPOpt, and instantiate it on the joint design of a cut selector and a branching rule. Candidate programs are iteratively generated, loaded into SCIP, and evaluated by direct execution on MILP instances, with the resulting feedback guiding performance-based selection, targeted repair, diagnostic reflection, and diversity-aware population maintenance. The method outputs explicit solver components that can be inspected, modified, and deployed within standard solver workflows. Across four benchmark families, we find that LLM-guided program evolution can discover competitive domain-specialized policies in several settings.",
                "url": "https://arxiv.org/abs/2607.18252",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "e2cae12b843b6efe78b7b046a2a618f8db889f82",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Beyond Accuracy and Cost: Latency-Aware LLM Query Routing for Dynamic Workloads",
            "summary": "arXiv:2607.18253v1 Announce Type: new Abstract: Modern language query routers improve inference efficiency by assigning each query to a model that balances response quality and monetary cost. However, current query routers are largely latency-agnostic and do not consider the generation latency experienced by queries at model instances. In practice, latency is often controlled by load-balancing policies such as round-robin or join-the-shortest-queue, which do not account for model accuracy or inference cost. Incorporating query latency into routing is challenging as it depends not only on the query's prompt length, but also on the current prefill and decode workload at the model instance and the scheduling and batching policy of the serving framework. We design a lightweight latency estimator that simulates autoregressive token batch processing in the serving framework and estimates the …",
            "url": "https://arxiv.org/abs/2607.18253",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Beyond Accuracy and Cost: Latency-Aware LLM Query Routing for Dynamic Workloads",
                "summary": "arXiv:2607.18253v1 Announce Type: new \nAbstract: Modern language query routers improve inference efficiency by assigning each query to a model that balances response quality and monetary cost. However, current query routers are largely latency-agnostic and do not consider the generation latency experienced by queries at model instances. In practice, latency is often controlled by load-balancing policies such as round-robin or join-the-shortest-queue, which do not account for model accuracy or inference cost. Incorporating query latency into routing is challenging as it depends not only on the query's prompt length, but also on the current prefill and decode workload at the model instance and the scheduling and batching policy of the serving framework. We design a lightweight latency estimator that simulates autoregressive token batch processing in the serving framework and estimates the time-to-first-token (TTFT) of queries. We incorporate this latency estimator into a latency-aware router that jointly optimizes latency, accuracy, and cost when assigning queries to model instances. Our experimental results indicate that this joint optimization yields up to 40% improvement in accuracy--cost utility while maintaining the same latencies as standard load-balancing approaches.",
                "url": "https://arxiv.org/abs/2607.18253",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "bc0c2a0779cddb120bb1c74bf5a0cc9dfe8dafd2",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Cross-Dialect Generalization Without Retraining: Benchmarks and Evaluation of Schema-Derived Constrained Decoding for MLIR",
            "summary": "arXiv:2607.18254v1 Announce Type: new Abstract: Multi-Level Intermediate Representation (MLIR) underlies modern ML compiler infrastructure (TensorFlow, JAX/StableHLO, PyTorch Inductor, IREE), yet appears only in trace amounts in code-LM pretraining corpora. MLIR is also extensible by design: new dialects ship per application domain, so a fine-tuned model per dialect does not scale. We ask whether inference-time priors derived mechanically from each dialect's Operation Definition Specification (ODS) can substitute for gradient-based adaptation. First, we release four natural-language-to-MLIR benchmarks across three dialects - MLIR-Spec-150, Linalg-Spec-30, StableHLO-Spec-30, and StableHLO-Held-Out-200 - totaling 410 in-scope NL-to-MLIR pairs, plus a 25-program out-of-grammar stress set and a hand-authored n=30 functional reference set, shipped under Apache-2.0 with Gebru datasheets and C…",
            "url": "https://arxiv.org/abs/2607.18254",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Cross-Dialect Generalization Without Retraining: Benchmarks and Evaluation of Schema-Derived Constrained Decoding for MLIR",
                "summary": "arXiv:2607.18254v1 Announce Type: new \nAbstract: Multi-Level Intermediate Representation (MLIR) underlies modern ML compiler infrastructure (TensorFlow, JAX/StableHLO, PyTorch Inductor, IREE), yet appears only in trace amounts in code-LM pretraining corpora. MLIR is also extensible by design: new dialects ship per application domain, so a fine-tuned model per dialect does not scale. We ask whether inference-time priors derived mechanically from each dialect's Operation Definition Specification (ODS) can substitute for gradient-based adaptation. First, we release four natural-language-to-MLIR benchmarks across three dialects - MLIR-Spec-150, Linalg-Spec-30, StableHLO-Spec-30, and StableHLO-Held-Out-200 - totaling 410 in-scope NL-to-MLIR pairs, plus a 25-program out-of-grammar stress set and a hand-authored n=30 functional reference set, shipped under Apache-2.0 with Gebru datasheets and Croissant 1.0 metadata. Second, we build a three-layer schema-derived constraint stack: a CFG over op signatures(C1), type-domain splits from an ODS-extracted type lattice (C2), and an SSA-scope validator driving five-retry rejection sampling (C3). Porting from arith+func+memref+linalg to StableHLO required no new constraint-layer code. On dialects whose verifier semantics are dominated by structural constraints, schema-derived priors let SmolLM2-1.7B match or exceed 15B-34B code LMs at 8-25x the per-generation speed: on linalg, SmolLM2 reaches 80.0% verify-valid (three-seed mean, n=125), beating CodeLlama-34B, Granite-Code-34B, and StarCoder2-15B by 21-44 percentage points with non-overlapping CIs. On arith+func and on the templated parametric StableHLO-Held-Out-200, where verifier semantics turn on attribute values rather than structure, the same baselines match or beat the SLM; we scope these as non-win cells. We release benchmarks, decoder, all per-prompt generations, and a reproducibility Docker image.",
                "url": "https://arxiv.org/abs/2607.18254",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "5aa67b3b90dfdb6beff8c2366e82f755e2c8eb27",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Semantic Cooperative Games for Contribution Attribution in LLM-Based Multi-Agent Systems",
            "summary": "arXiv:2607.18255v1 Announce Type: new Abstract: Contribution attribution has become a central problem in LLM-based multi-agent systems, where final outputs are produced through multiple agents, message exchanges, and ordered workflow dependencies. Existing attribution methods often rely on counterfactual valuation, such as removing agents or comparing score changes across altered agent subsets. In language-mediated workflows, these methods require repeated model calls, introduce high variance, and do not explicitly capture the intermediate semantic states through which agents produce, preserve, and transform task-relevant information. We propose Semantic Cooperative Games (SCG), a framework that represents a realized language flow as a semantic generation hypergraph and induces an agent-level semantic value function on this structure. We define the Semantic Shapley Value (SSV) to alloca…",
            "url": "https://arxiv.org/abs/2607.18255",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Semantic Cooperative Games for Contribution Attribution in LLM-Based Multi-Agent Systems",
                "summary": "arXiv:2607.18255v1 Announce Type: new \nAbstract: Contribution attribution has become a central problem in LLM-based multi-agent systems, where final outputs are produced through multiple agents, message exchanges, and ordered workflow dependencies. Existing attribution methods often rely on counterfactual valuation, such as removing agents or comparing score changes across altered agent subsets. In language-mediated workflows, these methods require repeated model calls, introduce high variance, and do not explicitly capture the intermediate semantic states through which agents produce, preserve, and transform task-relevant information. We propose Semantic Cooperative Games (SCG), a framework that represents a realized language flow as a semantic generation hypergraph and induces an agent-level semantic value function on this structure. We define the Semantic Shapley Value (SSV) to allocate contribution over semantic support logic, and introduce SLIC, a single-trajectory algorithm that constructs the semantic hypergraph, recovers minimal semantic supports, applies Boolean absorption, and computes SSV without rerunning agent subsets. We prove that SSV reduces to the classical Shapley value under standard set-based, fully observable, and no-order-dependence conditions. On a medical benchmark satisfying these conditions, SLIC reduces computation cost by 93.3% while remaining highly consistent with a Monte Carlo Shapley baseline. In more general multi-role workflows, SSV aligns with perturbation-induced score-drop profiles and exposes cases where semantic contribution and failure impact diverge. Overall, SLIC provides a fast, counterfactual-free, and interpretable attribution method for complex LLM-based multi-agent systems.",
                "url": "https://arxiv.org/abs/2607.18255",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "33d58d671b35c6453095fef49d741ebf4c8f87ca",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "PEARL: Solver-in-the-Loop Interactive Optimization Modeling from Natural Language",
            "summary": "arXiv:2607.18256v1 Announce Type: new Abstract: Optimization modeling is the process of translating real-world decision problems, often described in natural language, into formal mathematical formulations and executable solver code. While recent advances in large language models have shown promise in automating this process, most existing approaches remain one-shot: a model produces a formulation once, without executing it, conditioning on solver feedback, or iteratively revising errors. This stands in sharp contrast to real-world optimization modeling, which is inherently interactive and proceeds through repeated solve-debug-revise cycles. We introduce PEARL, a system for interactive optimization modeling that uses Python execution and mathematical programming solvers inside this loop. Rather than relying on a fixed repair workflow, PEARL learns when to test partial models, how to revi…",
            "url": "https://arxiv.org/abs/2607.18256",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "PEARL: Solver-in-the-Loop Interactive Optimization Modeling from Natural Language",
                "summary": "arXiv:2607.18256v1 Announce Type: new \nAbstract: Optimization modeling is the process of translating real-world decision problems, often described in natural language, into formal mathematical formulations and executable solver code. While recent advances in large language models have shown promise in automating this process, most existing approaches remain one-shot: a model produces a formulation once, without executing it, conditioning on solver feedback, or iteratively revising errors. This stands in sharp contrast to real-world optimization modeling, which is inherently interactive and proceeds through repeated solve-debug-revise cycles. We introduce PEARL, a system for interactive optimization modeling that uses Python execution and mathematical programming solvers inside this loop. Rather than relying on a fixed repair workflow, PEARL learns when to test partial models, how to revise from solver diagnostics, and when to stop. It operates in a multi-turn tool-integrated setting where intermediate execution results, feasibility signals, and solution checks are used to improve both formulations and solver code before finalization. Across diverse optimization benchmarks, PEARL substantially improves verified solve rates over strong one-shot and tool-augmented baselines; notably, our PEARL-Qwen3-\\textbf{4B} model outperforms the much larger DeepSeek-V3.2-\\textbf{685B} in both macro- and micro-averaged accuracy on optimization modeling tasks.",
                "url": "https://arxiv.org/abs/2607.18256",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "e33f292ee8c95d96082c08097c55b0492519b787",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "S2T-RLHF: Hierarchical Credit Assignment for Stable Preference-Based RLHF",
            "summary": "arXiv:2607.18258v1 Announce Type: new Abstract: Reinforcement learning from human feedback (RLHF) with preference-based reward models often exhibits unstable training dynamics. A key contributing factor is that standard RLHF relies on a single sequence-level scalar reward, which is propagated to token-level policy updates and leaves credit assignment within a response inherently ambiguous. Recent work has attempted to address this issue by refining rewards into denser token-level supervision, often relying on the implicit assumption that finer-grained credit assignment improves optimization. We argue that this assumption is incomplete: when preference signals are noisy and only defined at the response level, overly fine-grained reward refinement can amplify reward uncertainty and destabilize learning. To address this problem, we propose a granularity-aware principle for hierarchical cre…",
            "url": "https://arxiv.org/abs/2607.18258",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "S2T-RLHF: Hierarchical Credit Assignment for Stable Preference-Based RLHF",
                "summary": "arXiv:2607.18258v1 Announce Type: new \nAbstract: Reinforcement learning from human feedback (RLHF) with preference-based reward models often exhibits unstable training dynamics. A key contributing factor is that standard RLHF relies on a single sequence-level scalar reward, which is propagated to token-level policy updates and leaves credit assignment within a response inherently ambiguous. Recent work has attempted to address this issue by refining rewards into denser token-level supervision, often relying on the implicit assumption that finer-grained credit assignment improves optimization. We argue that this assumption is incomplete: when preference signals are noisy and only defined at the response level, overly fine-grained reward refinement can amplify reward uncertainty and destabilize learning. To address this problem, we propose a granularity-aware principle for hierarchical credit assignment, emphasizing stability-oriented reward design rather than maximal allocation precision. Under this principle, sentences serve as a natural intermediate granularity, balancing semantic coherence with robustness to token-level noise. Guided by this view, we introduce S2T-RLHF. This sentence-to-token reward decomposition framework first allocates sequence-level preference rewards across sentences and then applies bounded token-level refinement within each sentence, without reward-model retraining or token-level supervision. Experiments across multiple datasets and optimization settings show that S2T-RLHF improves training stability and robustness while maintaining competitive preference alignment.",
                "url": "https://arxiv.org/abs/2607.18258",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "6f47e167ebf2ddd7c1fe491c463c624b3767a746",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Probabilistic Concept-Aware Steering for Trustworthy LLM Inference",
            "summary": "arXiv:2607.18259v1 Announce Type: new Abstract: Steering vectors (SVs), an inference-time intervention technique for large language models (LLMs), guide the generation process by adding a concept-specific direction vector to intermediate activations during inference. However, existing SV methods frequently yield representation-incoherent behaviors that undermine interpretability and fine-grained control, largely because prior work has focused on binary positive-negative steering evaluation while employing discrete clustering metrics that fail to capture the continuous spectrum of semantic alignment. In this work, we present the Probabilistic Concept-Aware Steering (PCS) framework for LLM inference. PCS preserves original task competence while providing controllable, safety-oriented semantic bias through concept-driven steering-vector retrieval and probabilistic strength calibration.",
            "url": "https://arxiv.org/abs/2607.18259",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Probabilistic Concept-Aware Steering for Trustworthy LLM Inference",
                "summary": "arXiv:2607.18259v1 Announce Type: new \nAbstract: Steering vectors (SVs), an inference-time intervention technique for large language models (LLMs), guide the generation process by adding a concept-specific direction vector to intermediate activations during inference. However, existing SV methods frequently yield representation-incoherent behaviors that undermine interpretability and fine-grained control, largely because prior work has focused on binary positive-negative steering evaluation while employing discrete clustering metrics that fail to capture the continuous spectrum of semantic alignment. In this work, we present the Probabilistic Concept-Aware Steering (PCS) framework for LLM inference. PCS preserves original task competence while providing controllable, safety-oriented semantic bias through concept-driven steering-vector retrieval and probabilistic strength calibration.",
                "url": "https://arxiv.org/abs/2607.18259",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "e231121dfba922311bbb7fe900167cfb322f3f07",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "FindStatBench: Evaluating Large Language Models on Combinatorial Code Synthesis",
            "summary": "arXiv:2607.18260v1 Announce Type: new Abstract: We introduce FindStatBench, an execution benchmark for evaluating large language models on combinatorial code synthesis. Built from FindStat, it contains 2,329 tasks across 24 collections and 5.52M hidden instances, covering statistic synthesis, which maps objects to integers, and map synthesis, which maps objects to objects. Each task gives a mathematical description and at most five public input-output examples; a model must emit one Python solve function with no retrieval, tool use, execution feedback, voting, or reranking. Submissions are scored by exact sandboxed execution on held-out combinatorial objects. We evaluate eleven systems: four closed-source production models and seven open-weight models served through one inference provider. FindStatBench reveals three main patterns. First, the strongest open- and closed-source systems co…",
            "url": "https://arxiv.org/abs/2607.18260",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "FindStatBench: Evaluating Large Language Models on Combinatorial Code Synthesis",
                "summary": "arXiv:2607.18260v1 Announce Type: new \nAbstract: We introduce FindStatBench, an execution benchmark for evaluating large language models on combinatorial code synthesis. Built from FindStat, it contains 2,329 tasks across 24 collections and 5.52M hidden instances, covering statistic synthesis, which maps objects to integers, and map synthesis, which maps objects to objects. Each task gives a mathematical description and at most five public input-output examples; a model must emit one Python solve function with no retrieval, tool use, execution feedback, voting, or reranking. Submissions are scored by exact sandboxed execution on held-out combinatorial objects. We evaluate eleven systems: four closed-source production models and seven open-weight models served through one inference provider. FindStatBench reveals three main patterns. First, the strongest open- and closed-source systems converge within 1 pp instance accuracy, and both an oracle over all systems and five-way sampling from one mid-tier model yield only limited task-accuracy gains. Second, examples can hurt: several classical bijections are solved perfectly with zero examples but fail under five-example prompts. Third, some failures reflect output-budget mechanics, as reasoning can exhaust the visible response before code is emitted. Overall, statistic synthesis is much easier than map synthesis, some collections remain near-zero, long prompts cause a sharp accuracy cliff, and exact symbolic rule induction remains brittle.",
                "url": "https://arxiv.org/abs/2607.18260",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "755cacda77b736da925bc14706abc397a4880d75",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "When JSON Is Not Enough: Semantic Reliability of Schema-Constrained LLM Ordering Agents",
            "summary": "arXiv:2607.18261v1 Announce Type: new Abstract: LLM agents are increasingly used as transaction compilers: a user states an intent in natural language, and the model emits a structured object that an API can execute. JSON Schema and provider-level structured-output modes are useful because they remove a large class of parse failures, but they do not by themselves decide whether the object is a safe, faithful transaction. We introduce OrderBench, a deterministic benchmark for restaurant ordering agents that separates syntactic validity, schema validity, status decisions, exact item semantics, constraint preservation, and unsafe acceptances. Across 2,400 Nebius Token Factory calls to four open models in prompt-only and JSON-schema modes, we find that schema-valid output can still have large semantic error rates. In the strongest model, both modes achieve 100% schema validity, yet semantic…",
            "url": "https://arxiv.org/abs/2607.18261",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "When JSON Is Not Enough: Semantic Reliability of Schema-Constrained LLM Ordering Agents",
                "summary": "arXiv:2607.18261v1 Announce Type: new \nAbstract: LLM agents are increasingly used as transaction compilers: a user states an intent in natural language, and the model emits a structured object that an API can execute. JSON Schema and provider-level structured-output modes are useful because they remove a large class of parse failures, but they do not by themselves decide whether the object is a safe, faithful transaction. We introduce OrderBench, a deterministic benchmark for restaurant ordering agents that separates syntactic validity, schema validity, status decisions, exact item semantics, constraint preservation, and unsafe acceptances. Across 2,400 Nebius Token Factory calls to four open models in prompt-only and JSON-schema modes, we find that schema-valid output can still have large semantic error rates. In the strongest model, both modes achieve 100% schema validity, yet semantic success remains near 80%; in weaker models, schema-valid unsafe acceptances occur in double digits. The result is a concrete engineering warning: structured output is a necessary interface layer, not a substitute for domain verification and fail-closed execution.",
                "url": "https://arxiv.org/abs/2607.18261",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "e1412b0204ed30ecf39cd45c8fbdb0a7472efabc",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "ProbSPARQL: Querying Knowledge Graphs with Multi-dimensional, Uncertain Numeric Data",
            "summary": "arXiv:2607.18262v1 Announce Type: new Abstract: The SFB 1574 Circular Factory is building a shared knowledge graph infrastructure for integrating data about returned products. A central challenge is that circular-factory data include numeric measurements that (i) originate from sensors or are derived from sensor-based measurements, (ii) are frequently multi-dimensional, and (iii) are inherently uncertain, while downstream triage, validation, reliability-modeling, and reassembly-planning modules require queryable uncertainty representations. Current RDF and SPARQL technologies lack native support for harmonized querying and analysis of such uncertain numeric measurement data. To address this gap, we present ProbSPARQL, an upward-compatible SPARQL extension developed as an early-stage query-layer pilot for this infrastructure. ProbSPARQL models uncertain numeric values as random variables…",
            "url": "https://arxiv.org/abs/2607.18262",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "ProbSPARQL: Querying Knowledge Graphs with Multi-dimensional, Uncertain Numeric Data",
                "summary": "arXiv:2607.18262v1 Announce Type: new \nAbstract: The SFB 1574 Circular Factory is building a shared knowledge graph infrastructure for integrating data about returned products. A central challenge is that circular-factory data include numeric measurements that (i) originate from sensors or are derived from sensor-based measurements, (ii) are frequently multi-dimensional, and (iii) are inherently uncertain, while downstream triage, validation, reliability-modeling, and reassembly-planning modules require queryable uncertainty representations. Current RDF and SPARQL technologies lack native support for harmonized querying and analysis of such uncertain numeric measurement data. To address this gap, we present ProbSPARQL, an upward-compatible SPARQL extension developed as an early-stage query-layer pilot for this infrastructure. ProbSPARQL models uncertain numeric values as random variables whose distributions are encoded by probabilistic RDF literal datatypes, and supports distribution-aware expressions, probabilistic filters, and divergence-based joins. We implement ProbSPARQL on Apache Jena ARQ and expose it through a Fuseki-compatible execution layer. We assess real-data applicability using project-derived measurement fragments covering GMM-encoded uncertainty and histogram-based empirical roughness distributions, and evaluate scalability separately on controlled ontology-conformant benchmarks with up to 5,000 angle-grinder instances and 1.5M triples. The results show feasible in-engine execution, filter-pushdown speedups over application-layer post-processing, and latency-accuracy trade-offs among divergence-join decision strategies.",
                "url": "https://arxiv.org/abs/2607.18262",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "076632f4acf2d149f435e5ca78e364e3d05fb830",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Position: AI/ML Deepfake Research is Misaligned with AI-Generated Non-Consensual Intimate Imagery (AIG-NCII)",
            "summary": "arXiv:2607.18263v1 Announce Type: new Abstract: AI-generated non-consensual intimate imagery (AIG-NCII) is not adequately addressed in AI/ML literature regarding AI-generated media, commonly referred to as \"deepfakes\". While research on deepfakes currently focuses on its epistemic harms -- or harms relating to truth and authenticity -- this is misaligned with the dominant reality of generative AI abuse involving sexualized imagery. We conduct a landscape analysis of highly-cited works to demonstrate that technical interventions addressing deepfakes almost entirely ignore AIG-NCII, limiting the research ecosystem to authenticity detection tools. In this position paper, we argue that existing interventions address viewer-centric epistemic harms, such as fraud or scams, but ignore subject-centric dignity harms, such as AIG-NCII. We illustrate that knowing an image is synthetic does not mit…",
            "url": "https://arxiv.org/abs/2607.18263",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "Position: AI/ML Deepfake Research is Misaligned with AI-Generated Non-Consensual Intimate Imagery (AIG-NCII)",
                "summary": "arXiv:2607.18263v1 Announce Type: new \nAbstract: AI-generated non-consensual intimate imagery (AIG-NCII) is not adequately addressed in AI/ML literature regarding AI-generated media, commonly referred to as \"deepfakes\". While research on deepfakes currently focuses on its epistemic harms -- or harms relating to truth and authenticity -- this is misaligned with the dominant reality of generative AI abuse involving sexualized imagery. We conduct a landscape analysis of highly-cited works to demonstrate that technical interventions addressing deepfakes almost entirely ignore AIG-NCII, limiting the research ecosystem to authenticity detection tools. In this position paper, we argue that existing interventions address viewer-centric epistemic harms, such as fraud or scams, but ignore subject-centric dignity harms, such as AIG-NCII. We illustrate that knowing an image is synthetic does not mitigate harms to subjects and may, in some cases, even exacerbate them. We conclude by offering recommendations to realign the field, including updating threat models to consider subject-centric harms and addressing AIG-NCII in AI safety research. Finally, we caution that researchers should only engage in this high-risk domain if they implement safety guardrails for both subjects and researchers and establish partnerships with domain experts in sexual violence prevention.",
                "url": "https://arxiv.org/abs/2607.18263",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "c6d1819e0872089273bd5298a3e651c22a645c62",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "MUX: Continuous Reasoning via Multiplexed Tokens",
            "summary": "arXiv:2607.18264v1 Announce Type: new Abstract: Language models solve complex problems by articulating intermediate reasoning steps in natural language. While effective, this process is computationally bottlenecked: each reasoning step conveys only a single subword, and many are spent expressing a thought instead of carrying out computation. We propose MUX, a simple method for high-bandwidth and compact reasoning based on distillation of discrete reasoning into continuous multiplexed tokens in a latent space. Here, each latent token is trained to represent a weighted linear superposition (multiplexing) of a span of discrete reasoning subwords, where this superposition is lossless by construction and the span can be fully recovered (demultiplexing). We prove that simple position-dependent weightings, such as suitable geometric decay, support lossless multiplexing, which in turn prevents …",
            "url": "https://arxiv.org/abs/2607.18264",
            "image": "",
            "published": "2026-07-22T04:00:00+00:00",
            "score": 76.32,
            "color": "#b09cff",
            "raw": {
                "title": "MUX: Continuous Reasoning via Multiplexed Tokens",
                "summary": "arXiv:2607.18264v1 Announce Type: new \nAbstract: Language models solve complex problems by articulating intermediate reasoning steps in natural language. While effective, this process is computationally bottlenecked: each reasoning step conveys only a single subword, and many are spent expressing a thought instead of carrying out computation. We propose MUX, a simple method for high-bandwidth and compact reasoning based on distillation of discrete reasoning into continuous multiplexed tokens in a latent space. Here, each latent token is trained to represent a weighted linear superposition (multiplexing) of a span of discrete reasoning subwords, where this superposition is lossless by construction and the span can be fully recovered (demultiplexing). We prove that simple position-dependent weightings, such as suitable geometric decay, support lossless multiplexing, which in turn prevents shortcut behaviors caused by latent collapse. We further show that multiplexed reasoning can perform parallel exploration in problems that require search. Across 32 evaluation settings spanning four language models, MUX outperforms strong latent reasoning baselines. Ablation and probing analyses further show that the learned latent tokens encode faithful and interpretable reasoning. Our results suggest that lossless superposition as local learning targets constitutes a sufficient condition for achieving strong and efficient latent continuous reasoning.",
                "url": "https://arxiv.org/abs/2607.18264",
                "image": "",
                "published": "Wed, 22 Jul 2026 00:00:00 -0400"
            }
        }
    ],
    "stored": 0,
    "ai": {
        "enriched": false,
        "stored": 0
    },
    "health": [
        {
            "id": "arxiv_ai",
            "name": "arXiv — Artificial Intelligence",
            "engine": "rss",
            "category": "ai",
            "ok": true,
            "error": "",
            "ms": 236,
            "count": 20,
            "cached_at": "2026-07-23T01:58:07+00:00"
        }
    ]
}