arXiv:2611.00447v1 Announce Type: new Abstract: LLM-as-a-judge scores are often distributionally shifted from human ratings on any given evaluation task, and post-hoc calibration (mapping LLM to human scores in a small labeled anchor set) has become a popular method to correct for this shift. In this paper we derive the calibration from two positions: that LLMaJ scores are ordinal ratings drawn from a distribution rather than isolated points. We present streaming quantile matching (SQM), a versi
arXiv:2611.00460v1 Announce Type: new Abstract: Large language models (LLMs) now answer medical exam questions at or above expert level, but it is unclear whether such scores reflect clinical competence, or are instead affected by the format of the questions being asked (e.g., multiple-choice versus open-ended), prior exposure to the benchmarks, and the difficulty of the questions. These benchmarks are also public and widely used online, which may be resulting in LLM memorization. Models may per
arXiv:2611.00463v1 Announce Type: new Abstract: Agent harnesses govern how language models act in an environment; designing their policies by hand is difficult, while automated techniques require representative datasets and costly executions. We present Exo, an agent runtime architecture for safe learning during ongoing work that collapses the optimization loop into task execution, allowing agents to revise both their harness and their strategies for further improvement. A minimal protected mech
arXiv:2611.00464v1 Announce Type: new Abstract: Deploying small language models on resource-constrained hardware requires compression, yet practitioners lack systematic guidance on how quantization, pruning, and post-compression recovery interact when combined. We present a factorial ablation study evaluating 72 compression configurations across four SLM architectures (2.7B-7B parameters), extended with a 48-configuration ordering analysis, benchmark accuracy validation on 30 compressed models,
arXiv:2611.00474v1 Announce Type: new Abstract: Evaluating a stochastic AI system often requires many repeated samples per instance, especially for metrics defined over long horizons. Recent work has shown that modeling instance-level success probabilities with a Beta distribution recovers Pass@$k$ from far fewer samples than direct estimation. We show that this framework supports a broader range of evaluation targets. We fit a mixture of Beta distributions by marginal maximum likelihood, with t
arXiv:2611.00496v1 Announce Type: new Abstract: System One decision models replace text generation with typed, probability-scored answers and are entering production. Answers execute without a person in the loop, and no multi-model audit exists on established bias benchmarks. We provide that audit: three benchmarks: BBQ (all eleven categories), CrowS-Pairs, and SilverSpoon, adapted to the typed interface, twelve models plus a no-thinking ablation, and 1,638,000 decisions. Every order-sensitive i
arXiv:2611.00504v1 Announce Type: new Abstract: LLMs are increasingly used to design heuristics for computer systems. However, most synthesis frameworks evaluate candidates by their average performance, which can mask severe regressions on specific workloads. Even adding adversarial training inputs doesn't eliminate these regressions, and percentile objectives (e.g., optimizing for the 95th percentile) overlook how severe regressions can be beyond the selected threshold. We present TAILOR, a fra
Introducing LifeVal, a benchmark measuring how well AI helps people complete everyday tasks, from finding childcare to managing finances and government paperwork.
arXiv:2610.40758v1 Announce Type: new Abstract: Self-evolving agents rewrite a persistent state from feedback, but it is largely untested whether this lets them learn a rule nobody told them. An agent can also pass by writing down each correction as an exception, and its list then grows with every new rule. We introduce CLR-Bench, a Curriculum of Latent Rules benchmark. Each chain adds one hidden rule per level on top of the earlier ones. A rule acts on an opaque label that stands for a named gr
arXiv:2610.40798v1 Announce Type: new Abstract: Web world models increasingly predict the consequences of candidate actions before agents act, yet a plausible predicted future does not necessarily support the correct safety decision. In a matched human evaluation of 3,000 predictions from ten frozen open models, 86.1% are judged semantically plausible. However, several models exhibit highly asymmetric authorisation behaviour, combining low false-safe execution with poor recall of clarifiable ASK
arXiv:2610.40831v1 Announce Type: new Abstract: AI model marketplaces host providers with heterogeneous costs and capabilities. These differences create an opportunity for computational arbitrage (Olmedo et al., 2026), in which an intermediary purchases inference from a portfolio of models and resells it to users at a profit. Beyond creating profit opportunities for the arbitrageur, computational arbitrage offers financial value to both consumers and providers and could therefore significantly i
arXiv:2610.40883v1 Announce Type: new Abstract: Voice-interactive large language models process both what a speaker says and how they say it, yet the two channels do not always agree. We ask whether that disagreement degrades speech emotion recognition. Mapping text and audio into a shared valence-arousal-dominance space, we convert the distance between them into a continuous, threshold-free incongruence score with a two-component Gaussian mixture, and validate the score against human judgement.
arXiv:2610.40890v1 Announce Type: new Abstract: Large language models (LLMs) can generate detailed guidance, functional code, and persuasive text at scale. As one of the most widely used consumer language model products as of 2026, ChatGPT gives broad public access to these capabilities, motivating concerns about its potential use in surveillance-related tasks. We present an exploratory case study of what potentially harmful surveillance assistance looks like in ChatGPT's public web interface. W
arXiv:2610.40906v1 Announce Type: new Abstract: Recent scientific agents build knowledge recursively by retrieving evidence, constructing an intermediate state, and judging whether that state contains sufficient information. These agents are evaluated end-to-end, usually without ground truth, so the effect of each design choice is unknown. Here, we examined two design axes: (1) what the recursion builds (the state of evidence) and (2) what controls it (the judge that decides when to stop). To me
arXiv:2610.40917v1 Announce Type: new Abstract: Large language models act only through the infrastructure that surrounds them. That infrastructure, the harness, determines what a model can accomplish and how it fails, yet it has no settled definition. To address this gap, we decompose the harness into eight functional components that can be specified and compared across systems, and we propose a portable specification framework that records which model ran in which harness. The record is needed
arXiv:2610.40927v1 Announce Type: new Abstract: Test-time scaling (TTS) improves large language model agents by sampling and selecting among multiple candidate trajectories, but common realizations assume that these trajectories can be executed in parallel. This assumption breaks in realistic deployments, where enterprise workspaces and cloud services are shared and persistent, costly or infeasible to replicate, and actions may introduce irreversible side effects or cross-trajectory interference
arXiv:2610.40938v1 Announce Type: new Abstract: Using customer conversations to train and evaluate financial language models raises confidentiality concerns because non-public financial facts can remain after direct identifiers are masked. Synthetic transcripts offer an alternative data source, but their discourse fidelity and residual confidentiality risks require assessment. We introduce DiRAFT (Discourse Realization from Aggregate First-order Transitions), which extracts financial concept gra
arXiv:2610.40939v1 Announce Type: new Abstract: Chain-of-thought (CoT) monitoring is a promising way to oversee increasingly capable agents. It succeeds in part because current models appear limited in their ability to control what they verbalize in their reasoning traces. We investigate the source of this limitation and link it to an entropy collapse in the first few tokens following the opening thinking tag. We exploit this mechanism to recover CoT controllability through two inference-time in
arXiv:2610.40940v1 Announce Type: new Abstract: Digital agents can now gather and organize information, reason over it, and use tools to complete complex tasks. Forecasting puts all of these abilities to the test: it requires integrating incomplete evidence, inferring causal drivers, and committing to calibrated probabilistic judgments. Existing benchmarks are poorly suited to measuring and improving this ability. Live benchmarks provide feedback only after events resolve, so every evaluation cy
arXiv:2610.40946v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning allows large language models (LLMs) to generate intermediate reasoning traces in natural language while solving complex tasks. These traces can support safety monitoring by exposing potentially harmful plans before actions are executed. In this work, we investigate the conditions under which LLMs infer the presence of CoT monitoring from direct and indirect environmental cues and examine how this inference affects t
arXiv:2610.40959v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for personal tasks, yet evaluating their safe use of personal context remains difficult because such contexts are private, fragmented, dynamic, and socially situated. We introduce P-ContextBench, a realistic, verifiable, and scalable benchmark for evaluating personal-context understanding. P-ContextBench constructs population-structured, life-story-grounded digital humans from demographic seeds ref
arXiv:2610.40970v1 Announce Type: new Abstract: Generative audio systems can produce large libraries of personalized bedtime content, but ordinary media objectives (retention, completion, and novelty) are poorly aligned with the goal of supporting a transition toward sleep. Direct evidence for narrated bedtime audio in adults is mixed, and a recent active-controlled trial did not support such audio as a standalone treatment for sleep disturbance. We translate this uncertain evidence into conserv
arXiv:2610.40985v1 Announce Type: new Abstract: Good decisions often require considering how outcomes will feel. Yet large language models (LLMs) increasingly enter decision workflows, despite little being known about how they map outcomes to feelings or use those mappings in choice. We tested how LLMs anticipated gambling or stock-trading outcomes would feel and whether they incorporated these ratings into choice. When outcomes were rated in isolation, ratings exhibited diminishing sensitivity
arXiv:2610.40992v1 Announce Type: new Abstract: Managing multiple chronic conditions through diet requires satisfying disease- and medication-specific requirements jointly across meals and days while preserving established eating habits. These requirements are public in clinical guidelines and drug labels, yet turning them into a feasible plan for one person remains difficult for language-model agents. In our dietitian-approved weekly planning evaluation built from NHANES, thirteen models, from
arXiv:2610.41018v1 Announce Type: new Abstract: Violet teaming has been proposed as a method to integrate technical, policy, and societal considerations into the evaluation of emerging technologies. Here, we document a pilot expert workshop used to explore violet teaming as a governance methodology for biodefense-relevant AI capabilities and describe observations from the exercise. The workshop used a hypothetical technology scenario across three sequential stages - capability development, misus
arXiv:2610.41025v1 Announce Type: new Abstract: When a language-model agent fails a multi-step task, can another model find the decisive mistake and fix it with a single action? We introduce ReplayRCA, a benchmark that scores such diagnoses by intervention rather than by agreement with annotations: the environment is restored to the state preceding the chosen step, the proposed action replaces the agent's own, and the same agent continues. The score, causal repair, is the continuation success ra
arXiv:2610.41040v1 Announce Type: new Abstract: LLM agents serve natural-language requests through tools, data sources, and other agents, and even simple requests often need several interdependent calls. Typically, the LLM picks these calls by trial and error, with no guarantee of an executable plan. We present Capability-Grounded Agentic Planning (CGAP), which instead reasons about what a request requires and what tools provide. CGAP compiles the tool catalog deterministically into a PDDL model
arXiv:2610.41044v1 Announce Type: new Abstract: The use of large language models (LLMs) to coordinate multi-agent teams raises a fundamental architectural question: whether task allocation should be centralized in a single LLM controller or distributed across autonomous LLM agents, particularly when communication is unreliable. In a deliberately simplified perimeter-defense simulation against LLM-controlled intruders, we compared a central commander, an auction-based consensus team and independe