overfeed.news
Anuncio

NexTalk

Automatiza la primera respuesta. Mantén el toque humano.

Chat en vivo con agentes de IA en tu modelo y tu clave de API, con traspaso a tu equipo.

Empezar gratis

Tema

LLMs

2993documentos

239últimos 7 días

arXiv AI Papers

ALLMTS: Automated LLM-based Medical Triage System

Telephone triage is a safety critical task: from a brief intake note, a nurse must decide whether a caller needs an ambulance, an emergency department, an appointment, or self care at home. Queried directly, large language models (LLMs) are unreliable: they over-escalate routine cases and cannot justify their recommendations against an auditable clinical standard. We present the Automated LLM-based Medical Triage System (ALLMTS), which grounds every disposition in the same commercially maintained telehealth guidelines nurses use. Rather than asking an LLM to triage, ALLMTS makes it execute the guideline: it selects the appropriate guideline and evaluates its triage assessment questions. The disposition follows deterministically from the first positive answer, using the same traversal rule a nurse follows. We evaluate ALLMTS on 566 synthetic, expert-validated scenarios spanning 14 dispositions and designed to test specific guideline questions. ALLMTS achieves 0.95 accuracy, compared with 0.47 for a raw-LLM baseline. On a curated 50-scenario subset released as an open benchmark, ALLMTS reaches 0.98 accuracy, on par with the 0.944 correct disposition rate that a five nurse expert panel recorded on those same scenarios. Critically, on that subset the baselines have over-triage rates of up to 0.50 (0.44 on the full dataset), whereas ALLMTS reduces the over-triage rate to at most 0.02 while maintaining low under-triage. These results support coupling LLM language understanding with executable clinical guidelines for triage from intake notes.

en
arXiv AI Papers

Sharpness-aware Orthogonalized Momentum Algorithms

Adversarial sharpness penalty per iteration of a base optimizer, known as sharpness-aware minimization (SAM), is an accepted technique to improve generalization in deep learning. Recently, Muon has emerged as a dominant optimization algorithm for training large language models (LLMs). It achieves lower sample complexity for heavy-tailed gradient updates - a phenomenon common in LLM training. However, sharpness penalties with Muon updates are yet largely unexplored in the literature. Moreover, the behavior of SAM under heavy-tailed noise is unknown both theoretically and empirically. A potential reason for this gap is the additional cost of extra gradient computation per iteration for penalty perturbations in SAM, which becomes prohibitive for LLMs. This paper introduces Sharpness-aware Orthogonalized Momentum Algorithm (SOMA) to combine a computationally efficient and theoretically grounded perturbation scheme with the Muon algorithm. The key idea is to use stale (purely) stochastic gradient noise for adversarial perturbations. Theoretically, we show that SOMA's sample complexity is invariant to the perturbation direction. We provide explorations and insights into SOMA's robustness to heavy-tailed gradient noise. We extensively explore SOMA's efficacy in (a) mitigating quality degradation due to post-training quantization, and (b) catastrophic forgetting during fine-tuning of LLMs.

en
arXiv AI Papers

Reading Between the Layers: Cross-Modal Alignment and Representational Hierarchy in Vision Transformers

While large vision-language models excel at optical character recognition, it remains unclear how their underlying vision encoders process text and its meaning, and whether they support anything resembling reading comprehension. We systematically study transformer-based vision encoders, using cross-modal alignment between rendered text and visual concepts together with probing and retrieval-based analyses as a controlled window into these broader capabilities. Our layer-wise analyses span orthographic, grammatical, and semantic organization, and reveal that strong cross-modal mapping emerges only in the deepest layers, after unimodal object and text recognition have already developed, coinciding with a progression from surface-level to semantic representations. Although this hierarchy is strongest in contrastively trained models, larger self-supervised models and text-rich training data partially support similar behaviors without explicit language supervision. Together, our results suggest that increasing model and training-data scale, as well as text-rich training data, can partially compensate for the absence of language supervision.

en
arXiv AI Papers

Decoupling Patient Personas and Clinical Workflows: A Dual-Agent Approach to Realistic Medical Dialogue Generation

Pre-consultation medical dialogues are essential for collecting comprehensive patient information, but are costly to obtain at scale due to privacy and ethical constraints. Existing LLM-based synthesis methods often rely on prompt engineering or rigid rules, limiting their ability to balance clinical relevance with realistic patient diversity. We propose CoRPS-Dial, a controllable dual-agent framework for pre-consultation medical dialogue synthesis that integrates patient personas with guideline-aligned clinical workflow rules. CoRPS-Dial generates dialogues through interactions between a persona-driven patient agent and a process-guided doctor agent, followed by post-hoc quality control for role consistency, workflow completeness, and textual consistency. Experiments on coronary heart disease demonstrate that CoRPS-Dial improves alignment between synthetic and real dialogues in information coverage, clinical compliance, and dialogue naturalness. Downstream fine-tuning experiments further confirm the utility of the generated synthetic data for medical dialogue tasks. These results highlight the effectiveness of CoRPS-Dial for clinically grounded and controllable medical dialogue synthesis.

en
arXiv AI Papers

SCI-AGENT-EXAM: Benchmarking Agents on Scientific Research Tasks

Large language model (LLM) agents are increasingly capable of supporting scientific workflows, but their capabilities across diverse research activities remain difficult to assess systematically. Existing benchmarks often focus on selected scientific capabilities or task types, limiting a comprehensive characterization of scientific agents' strengths and weaknesses in realistic research workflows. We introduce SCI-AGENT-EXAM, a workflow-oriented benchmark comprising 120 expert-curated tasks across 11 scientific fields and 7 research task types defined by the Scientific Workflow Task Taxonomy (SWTT). Each task requires both executable artifacts (e.g., code and scripts) and textual artifacts (e.g., reports and explanations). To evaluate the heterogeneous task outputs, we develop a hybrid evaluation protocol combining deterministic programmatic verification for executable artifacts with a calibrated, expert-validated LLM-based judge for textual artifacts. Beyond outcome assessment, we analyze progress and solution stability along execution trajectories to characterize how agents reach their final results. Experiments with 10 frontier LLM agents reveal substantial performance gaps: GPT-6 Astra achieves the highest overall score of 80.9, yet only 26.0% of its runs exceed the 0.95 success threshold. Further analyses uncover distinct capability profiles across scientific task types and substantial variation in execution efficiency and stability. Our code and data are available at https://github.com/SKYLENAGE-AI/Sci-Agent-Exam.

en
arXiv AI Papers

Sensorimotor Alignment: A Platonic Proxy for Embodied Policy Success

Robotic foundation models are typically evaluated through costly closed-loop rollouts, which motivates cheap offline proxies for model capability. The Platonic Representation Hypothesis (PRH) suggests that as representations improve, different modalities such as vision and language converge toward a shared statistical model of reality. We ask whether motion trajectories form another such modality: as embodied models become more capable, should their sensory representations converge toward the structure of physical motion? To examine this, we propose Sensorimotor Alignment (S^2), a training-free offline metric that measures how well a model's sensory representation aligns with the geometry of physical motion. We find that S^2 correlates positively with downstream robot task success across embodied model families, making it a low-cost offline proxy for prioritizing models before closed-loop evaluation. S^2 can also identify promising vision-language model (VLM) backbones even before costly fine-tuning into vision-language-action models (VLAs). Beyond this correlation, S^2 offers representation-level explanations for phenomena observed in VLAs and world action models (WAMs). With S^2, we extend PRH to embodied models and make sensorimotor alignment empirically measurable, offering insight into why an embodied model succeeds beyond its closed-loop success rate.

en
arXiv AI Papers

Reliable AI-Assisted Decisions via Selective Auditing

We consider online monitoring of an AI system that produces a sequence of decisions. At each round, the system outputs a proposed decision together with a confidence score estimating the trustworthiness of the decision. The decision and confidence score need not be produced by the same model: for example, one agent may propose a decision, while a separate verifier agent evaluates it and outputs a correctness or safety score. Based on this information, we must decide whether to accept the AI decision automatically or consult an expert, such as a human auditor, who reviews the proposal and either accepts or corrects it. We formulate this task as an online constrained optimization problem in which the goal is to minimize the frequency of expert auditing while guaranteeing that the fraction of AI errors that go unaudited stays below a user-defined threshold. We derive a family of algorithms with provable error-control guarantees with no assumptions, including a method that is optimally efficient: it asymptotically attains the best achievable auditing rate for both adversarial and i.i.d. data streams. Experiments on screening harmful requests to large language models, qualitative text annotation, and data labeling show that our methods achieve valid error control in practice while substantially reducing the need for human auditing compared to baselines.

en
arXiv AI Papers

Cost-Aware Abstention for LLM-Based IT Ticket Classification

Wrong automatic routing of an IT support ticket can cost more than human triage. We evaluate cost-aware abstention as an offline, cost-conditional readiness assessment: select a routing policy on calibration data under explicit costs, then evaluate that frozen policy once on held-out test data. On a public 47,837-ticket, eight-class corpus, we compare four trained non-LLM classifiers with four large language models under zero- and few-shot prompting. Under the illustrative cost model (c_ok, c_triage, c_wrong) = (0, 1, 5), TF-IDF with logistic regression and a threshold of 0.60 selected on calibration reduces test cost from 0.763 to 0.472, a 38.1% relative reduction (paired ratio-bootstrap 95% CI [35.7%, 40.4%]), while lowering wrong routes from 15.3% to 3.5%. Frozen calibration-selected policies also beat both always-route and always-defer on held-out test data in 14 of 16 settings in a post-hoc cost-parameter analysis. The strongest prompted configuration reaches 0.477 accuracy on a stratified 1,000-ticket test subset, versus 0.879 for fine-tuned DistilBERT on the same tickets. Isotonic lowers expected calibration error without reliably improving discrimination: GPT-4o-mini few-shot has one confidence value before and after recalibration and correctness AUROC 0.500. We therefore report correctness AUROC and tie-aware risk-coverage area alongside ECE. At the analytically derived break-even probability 0.80, seven prompted configurations defer all tickets; Haiku few-shot routes only 19/1,000. An LLM-only agreement control and margin/entropy rejection baselines further separate calibration from discrimination. The findings are scoped to this dataset, prompt design, model endpoints, and illustrative costs; the work is an evaluation framework, not a new abstention algorithm or universal deployment certification.

en
arXiv AI Papers

EpiSEM-Bench: Evaluating Safety and Evidence Recovery in LLM Agents under Missing Evidence

Tool-using language model agents rely on environmental facts to decide whether an action is permitted. When a tool response omits a safety-relevant fact, an agent may carry out a prohibited action even though the user request is benign and every returned statement is true. We introduce {}, an executable benchmark for evaluating unsafe execution, safety boundaries, and evidence recovery in LLM agents under missing evidence. It contains 300 matched unsafe/safe task pairs across six domains. By systematically suppressing critical facts, the benchmark measures unsafe execution under critical evidence omission and determines the minimum observed failure cardinality, including failures with full evidence and cases with no observed failure after exhaustive testing. An independent audit tool tests whether agents seek missing facts and then withhold prohibited actions while completing permissible tasks. The benchmark reveals substantial variation across the ten models in omission robustness, evidence acquisition, and appropriate action. Resistance to missing evidence does not necessarily translate into useful recovery. Recovery access reduces unsafe execution for all ten models, but agents differ in whether they actively obtain the missing facts and use them to act appropriately in both worlds.

en
arXiv AI Papers

Low-Bit N-Gram Embeddings for Language Models

Many recent open-weight language models use n-gram embedding tables to improve performance. These tables store learned representations of short token sequences, such as pairs or triples of consecutive tokens, which the model retrieves during processing. However, they can require substantial memory: DeepSeek-V4.1-Flash, for example, allocates 196 billion parameters to its Engram n-gram memory. We therefore investigate whether this memory cost can be reduced while preserving the performance improvement that the tables provide. We propose storing n-gram embeddings at low precision and systematically investigate different quantization methods and bit allocations across tables. Across nanochat-, Llama-, and GPT-2-family models trained from scratch, four-bit tables save approximately 75% of table memory while preserving more than 99% of the validation-loss improvement from adding the tables. Using fewer bits leads to a larger drop in performance, although we find that a short fine-tuning phase recovers a significant part of this loss. In addition, reinvesting the saved memory in larger tables can improve performance beyond that of the original full-precision tables while still using less memory. These table memory savings also extend to large existing models: on DeepSeek-V4.1-Flash and Qwen3.8-Flash-Next, low-bit tables save 47-55% of table memory, with perplexity increases of less than 1.2% and largely preserved downstream task performance. These findings establish low-bit precision as a practical way to reduce the memory requirements of n-gram embeddings and use the saved capacity to improve model performance.

en
arXiv AI Papers

Can an Agent Debug a Web Corpus? Agentic Auditing of Slovak FineWeb2

Multilingual pipelines such as FineWeb2 apply one recipe across a thousand languages, leaving the failure modes of individual languages unanticipated. We use an LLM once, as a corpus debugger rather than a classifier: a tool-using agent explores the corpus through a read-only, budgeted API and compiles each finding into an executable test, which a native speaker then validates. On the 26.5M-document Slovak portion of FineWeb2, three runs costing \2.21-\2.84 each found seven findings over four phenomena. The largest is a machine-generated content farm of 1,408,437 documents (5.32%) that passed the pipeline's filters, 86.6% of it passing even stricter thresholds. Blinded labels support 6 of the 7 findings (defect rates .53-1.00 against .08 in a random control) and do not support one. The compiled suite scans the full corpus in 0.7 s of wall-clock time without a GPU, against \12k-\90k per LLM classification pass. The best static sampling baseline validates more selectors than any agent run, at under half the cost. The agent contributes exact prevalence, a precision estimate checked against sampled documents, an account of why the pipeline missed each slice, and a reusable test. We release the framework, Slovak test suite, validation data and agent trajectories.

en
arXiv AI Papers

Bilingual Language Models Capture Cross-Linguistic Influence in Second-Language Reading

While monolingual language model (LM) surprisal is widely established as a predictor of reading times, it cannot capture the non-native (L2) reading dynamics that are shaped by the reader's first language (L1). In this work, we investigate whether the cross-lingual transfer within bilingual models aligns with human L1-driven biases during L2 reading. We evaluate bilingual models on eye-tracking data of readers from 12 diverse L1 backgrounds reading English texts. To isolate L1-specific transfer, we compare bilingual models trained on an L1 that matches the readers' L1 background to both a monolingual English baseline and bilingual models trained on other L1s. We demonstrate that matching the readers' L1 improves the prediction of L2 reading times. This predictive advantage is concentrated in late-stage eye-tracking measures, modulated by L1-L2 script overlap, and peaks among intermediate-proficiency readers. These findings provide evidence that bilingual LMs capture aspects of cross-linguistic influence, and that such influence is present in L2 reading behavior.

en
arXiv AI Papers

How Ground Truth Shapes Retrieval Scores: Evidence-Aware Evaluation of a Retrieval-Augmented Generation System for Breast Lesions

Purpose: To describe an evidence-aware retrieval evaluation protocol for a clinical retrieval-augmented generation (RAG) system over breast lesion management literature, addressing the dependence of retrieval scores on annotation coverage. Methods: An 18-document corpus and an 18-query development benchmark, with one Yes/No or WH-question per document, were used. Retrieval combined dense embeddings, BM25, and cross-encoder reranking. Chunk sizes of 128 to 1024 tokens, overlaps of 0 to 20%, and four retriever configurations were evaluated. Relevance was defined by evidence spans mapped to all containing chunks, treating overlapping chunks with equivalent evidence as relevant. Retrieval was evaluated using R-Precision, Recall@K, MRR, and nDCG@K. Separately, generation was assessed on a 512-token index at K=1 to 5 using DeepEval claim-level metrics, with deepseek-v4-pro as the LLM judge, and BLEU/ROUGE metrics. Results: The selected hybrid retriever, using 384-token chunks and 20% overlap, retrieved accepted evidence in the top three results for 17 of 18 queries. Recall@3, R-Precision, MRR@3, and nDCG@3 each reached 0.94. Ground-truth curation increased evidence spans from 25 to 37 and accepted chunk targets from 36 to 51 without changing retrieval rankings. Conclusion: Cross-encoder reranking achieved the highest retrieval scores. Expanded relevance annotations increased recognized evidence without changing rankings, highlighting annotation sensitivity. The small development set and retrieval-informed curation require independent validation, including radiologist assessment of evidence sufficiency.

en
arXiv AI Papers

Simulated Control Evals

Control evaluations measure whether safeguards can prevent an AI model from causing undesired outcomes even when it is intentionally trying to do so. These evaluations typically use purpose-built control settings and extrapolate their results to a model's deployment. However, the effectiveness of safeguards in deployment can depend on features of environments and tasks that control settings do not represent. To broaden where control evaluations can be run, we introduce simulated control evaluations (SCEs), which use a language model to simulate tool responses in place of an environment. We evaluate SCEs on LinuxArena, where execution provides ground-truth outcomes. We compare SCE prediction errors with those from a separate control setting, BashArena, and with variation in control metrics within LinuxArena. For independently sampled evaluation and deployment task sets, SCEs predict safety with a root mean squared error of 3.6 percentage points across five monitors at a 5% audit budget, compared with 10.1 for BashArena. SCEs closely reproduce distributions of attack strategies, attacker reasoning, and monitor reasoning, with over tenfold lower KL divergence from execution than BashArena. SCEs explain 78-85% of the between-environment variance in attack success, monitor scores, and safety. This accuracy is sample efficient: SCEs predict safety with less error than BashArena using fewer than 5% as many task-pair draws as the deployment task set. Strategies discovered through simulation also improve monitor performance on held-out LinuxArena attacks, outperforming BashArena and approximately matching improvement from executed strategies. These results suggest that deployment-specific SCEs can be more predictive than control evaluations executed in separate control settings.

en
arXiv AI Papers

VERA: A Responsible AI Dashboard for Assessing LLM Deployments in Light of the EU AI Act

Institutions that deploy large language models in regulated settings often need technical evidence of responsible AI practices before approving a deployment. COMPL-AI provides a technical interpretation of the EU AI Act that organises this evidence under six principle groups and eighteen technical requirements, twelve of which current benchmarks can measure. The benchmark suites that produce this evidence are typically operated from a command line, while the staff who approve deployments are non-engineers. We present VERA, an open-source self-hosted dashboard deployed to facilitate responsible AI processes within a large European bank. A reviewer selects a model and runs public benchmarks against it. VERA scores the twelve requirements that current benchmarks can measure and instruments the energy consumed by each evaluation, adding run-level evidence for environmental impact. We run four served open-weight models through the requirement pipeline, and the dashboard displays a score for each measurable requirement. A within-subject study then compares how participants read these benchmark results: from the raw output files in one condition, and through the VERA dashboard in the other. The dashboard reduced the time needed to perform the responsible AI assessment, reducing the time required from 56 seconds using raw files to 30 seconds using our dashboard, on average. Our dashboard scored highly (4.36 out of 5) on usefulness in the survey conducted within the organization. The dashboard and the study material are released publicly to promote replicability and reproducibility in different organisational contexts.

en
arXiv AI Papers

Would an LLM Match into a Residency? Uncertainty, Reasoning and Image Use of Open Language Models on the Spanish MIR Examination

Accuracy on medical examinations is common evidence offered for the clinical use of language models, yet a health system that intends to run an open model on its own hardware needs to know considerably more: whether the model recognizes when it is likely to be wrong, whether medical fine-tuning improves this ability, whether its correct answers rest on sound clinical reasoning, or whether it actually uses the images it is given. We examine these questions for four open models of 8 to 27 billion parameters on the Spanish MIR residency examination, scoring 1,552 questions from eight editions under the exam's own rules. Answering every question, the three larger models would rank among roughly the first 600 to 3,000 candidates, with 41 to 43 specialties still open to them. Abstention, however, never pays, because even their least confident answers are right more often than the 25% break-even. Medical fine-tuning reduced overconfidence at the cost of accuracy, without improving how well confidence separates right from wrong answers. Moreover, 13-38% of correct answers rejected a distractor for a clinically unsound reason that confidence did not reflect, and removing or swapping the provided contextual images left accuracy unchanged.

en