Noticias de IA - ítems

arXiv:2610.33651v1 Announce Type: new Abstract: Do preferences Large Language Models (LLMs) express in standard elicitation tasks predict their strategic behaviour? LLM agents interacting with humans and other AI systems face potentially conflicting objectives, such as individual welfare, group welfare, and environmental concerns. Prior work characterizes LLM preferences through behavioral-economic and psychological elicitation or strategic games, but rarely tests consistency across settings and prediction in other strategic environments. This requires modeling how competing objectives jointly shape behavior, whereas existing approaches typically measure preference dimensions in isolation. Drawing on utility theory, we introduce the Preference-Behavior Model (PBM), which combines a preference representation with mechanisms capturing imperfect utility maximization. We formulate estimation as an inverse-game problem, jointly inferring preferences and behavioral parameters from observed choices. We evaluate PBM in three social dilemmas that expose conflicting motives: a Public Goods Game, a Collective-Risk Dilemma, and a Common-Pool Resource dilemma. We find that while preferences estimated from established behavioral economic elicitation instruments do not consistently predict behavior in these dilemmas, models fitted in one dilemma retain predictive value in others, suggesting that inferred preferences capture meaningful aspects of LLM decision-making that elicitation instruments may miss. Finally, we assess alignment with persona instructions, finding that inferred preferences broadly reflect intended priorities with specific divergences across dimensions, LLMs, and games being identified by our approach. Our methodology provides a framework for studying how consistently LLM preferences explain behavior, assessing alignment with intended priorities, and revealing trade-offs and behavioral patterns that standard elicitation may miss.

arXiv:2610.33650v1 Announce Type: new Abstract: Online distillation of vision-language-action (VLA) policies commonly relies on teacher action scores that an inference-only interface may not expose. Trajectory-only Adversarial Distillation (\gad{}) instead turns a fixed bank of teacher joint-position features into adaptive feedback: a discriminator compares feature marginals from teacher and student rollouts and supplies rewards for on-policy learning. The student interacts with the environment without online teacher queries or environment task rewards. On ManiSkill \texttt{PutOnPlateInScene25Main}, \gad{} improves OpenVLA-OFT (7B) from $26.2\%$ to $76.6\%$ \successonce{} ($+50.4$ percentage points) and from $20.3\%$ to $58.6\%$ \successend{}. It trails white-box on-policy distillation (OPD) at $81.6\%$ / $59.4\%$ by $5.0$ / $0.8$ percentage points ($n{=}256$ evaluation trajectories per checkpoint). The tested four-epoch continuation achieves the highest \gad{} result without additional teacher data. This best run uses $5\times$ OPD's student interaction budget and $12.5\times$ its complete-rollout batch-passes. The comparison demonstrates online VLA improvement without teacher scoring while quantifying unequal interaction and optimization budgets.

arXiv:2610.33644v1 Announce Type: new Abstract: Research ideas are always proposed under resource constraints: a plan must fit the available compute, API credit, and time. Yet evaluations of research-ideation agents mostly score unconstrained novelty or overlap with a published paper, and their LLM judges are rarely checked against peer review. We introduce AlphaIdeaBench, a benchmark of 100 ideation tasks (50 workshop calls for papers and 50 open problems stated in surveys), each paired with an explicit resource budget and a shared proposal schema. An LLM judge scores each proposal on 12 dimensions covering validity, novelty, and feasibility. To validate the judge, we compare its scores on four novelty axes (problem, method, scenario, and insight) with scores derived from reviewer comments on 600 papers from ICLR 2026, NeurIPS 2025, and ICML 2025. For problem and method novelty, Gemini 3.7 Flash is within one point of the reviewer-derived score on 94-97% of papers; agreement on scenario novelty is relatively lower. We evaluate eight published harnesses on a shared backbone and, within one harness, eight backbone LLMs. CoI-Agent obtains the highest overall score, narrowly ahead of SciMON, which leads on method and insight novelty. All harnesses score low on scenario novelty and rarely change the problem setting. The backbone largely determines which open problem a proposal targets, whereas the harness determines whether a proposal addresses one problem or combines several. Weaker backbones such as Llama 4 Scout fail to carry out that harness's strategy and score low on budget feasibility. Finally, when only the budget is varied on three topics, a larger budget changes the experimental setting on compute-intensive topics but only adds tests on an inexpensive one. We release the tasks, rubrics, and scores to foster standardized, reproducible benchmarking of future research-ideation agents under constraints.

arXiv:2610.33625v1 Announce Type: new Abstract: Multi-Head Latent Attention (MLA) has been adopted by recent frontier large language models. Instead of per-head keys and values, MLA caches a compressed KV latent that is up-projected into keys and values at decode time, together with a narrow positional key that carries position information into the attention score. Existing KV-cache quantization methods are designed around the keys and values of standard attention and do not account for what MLA actually caches, which costs accuracy at low bit widths. We analyze the MLA cache both in activation distribution and in how quantization error propagates to the attention output, and show that the two cached tensors play different roles: the latent mainly supplies the values, whereas the positional key dominates the attention score. We therefore propose RAQ-MLA (Role-Aware Quantization for Multi-Head Latent Attention Caches), which fits an invertible transform to each cached tensor against the error that tensor delivers to the attention output: attention-weighted value error for the latent and score perturbation for the positional key. Without training, RAQ-MLA quantizes the full MLA cache to about 2.5 bits per element, a 6.5x increase in theoretical KV-cache capacity, while remaining nearly lossless on reasoning and coding tasks across four MLA models.

arXiv:2610.33623v1 Announce Type: new Abstract: Graph Neural Networks are widely used to learn structural representations of relational data. Hierarchical graph pooling learns multi-level representations by progressively coarsening graphs; however, prior methods rely on predefined cluster numbers or fixed pooling ratios, limiting adaptability to individual graphs. We propose Graph-Dependent Pooling (GDPool), which dynamically determines pooling granularity for each graph. We analyze optimization conflicts among layer-wise objectives through gradient alignment, showing that independently supervised pooling levels can send conflicting gradients to shared parameters. GDPool uses self-attention to select graph-specific anchors representing structurally important nodes and constructs coarse graphs through soft node assignment. Experiments on metal-organic framework and molecular graph benchmarks show that GDPool consistently achieves the best predictive performance across all benchmarks. Evaluations using modularity and the effective number of clusters show that GDPool achieves consistently higher modularity than existing methods while the number of active clusters decreases monotonically with depth. Sensitivity analyses confirm stable behavior across hyperparameter and architectural choices.

arXiv:2610.33617v1 Announce Type: new Abstract: Large language models (LLMs) increasingly produce online articles from material that later becomes input to other models. This circulation connects individual responses to the information available for subsequent generation. We study a consequential distinction: an article can avoid endorsing a false claim while preserving the concrete details that enable its reproduction. Our framework connects claim content, expressed stance, and downstream reuse to explain selective error persistence under a continuing supply of accurate and inaccurate information. We implement a controlled community writing process that varies whether published articles return as inputs, matching external supply, input capacity, and generation volume. Across twelve factual targets and four production models, recirculation increases the presence of complete false claims by 7.29 percentage points at the final round while reducing explicit endorsement by 8.33 points. Effects differ across facts and models. Paired interventions then abstract eligible error details within fixed histories while preserving correct content and the surrounding controversy. Error reproduction remains largely unchanged across the evaluation targets, while a development quotation target exhibits local reductions. These findings connect persistent errors to the information available throughout a writing process and motivate evaluation of both what models affirm and what their publications supply to future writers.

arXiv:2610.33605v1 Announce Type: new Abstract: This book develops a unified governance framework for systems that retrieve, generate, infer, remember and act on information from large language models and retrieval augmented pipelines to autonomous multi agent systems. Eleven chapters and eight appendices integrate three disciplinary lenses-technology, law and economics-across the full lifecycle: data quality, lineage and provenance; global data protection law and intellectual property; the economics of data governance; digital trust, cybersecurity and zero-trust architecture; ethics and responsible AI; governing generative AI pipelines; governing agentic AI and multi-agent systems; data governance in elections and the public sphere; AI regulation across six jurisdictions; and the design of a governance control plane for 2030. Each chapter is anchored in Indian law and institutions while maintaining a comparative, cross jurisdictional perspective.

arXiv:2610.33594v1 Announce Type: new Abstract: Deployed machine learning (ML) models can degrade as the distribution of their input data shifts. Such shifts can result from changes in the population or from data integrity issues in the pipeline that materializes model features, two sources that operate through different mechanisms. Because labels are often absent or delayed, it is common practice to track shifts in the data instead, but a detected shift does not indicate how it affects the model. We formulate this problem as online model testing under distribution shift. Testing requires reproducing the shift on a controlled baseline with known labels. Since the same shift can come from either source, doing so requires knowing which mechanism changed. We propose a methodology that reproduces observed changes in both mechanisms to estimate current performance and the potential gain from repairing observed data integrity issues. Experiments on real-world datasets with simulated faults show that separating the two sources generally improves both estimates, supporting this formulation. A churn prediction case illustrates how these tests can support monitoring and repair prioritization

arXiv:2610.33592v1 Announce Type: new Abstract: Augmented large language model (LLM) systems have reported high benchmark accuracies in rare disease diagnosis, yet their real-world clinical utility remains unproven. A recent meta-analysis flagged augmented LLM diagnostic evaluations as having a high risk of bias due to suspected data leakage; however, the precise impact of these vulnerabilities on model inference has yet to be fully quantified. In this study, we present a comprehensive empirical audit to quantify evaluation artifacts across prevailing AI paradigms, using a repository of over 60,000 cases from diverse public and independent clinical cohorts. By reconstructing a state-of-the-art agentic framework, we demonstrate that candidate-space contamination artificially inflates performance: enforcing strict cross-dataset isolation resulted in a marked performance decline, with the cross-dataset mean Recall@1 decreasing from 54.3% (under conventional one-vs-rest evaluation) to 40.3% (Hedges' g = 0.94). Similar spatial overfitting in supervised classifiers caused diagnostic accuracy to drop from an internal mean of 42.9% to 16.7% across external cohorts (Cohen's d = 2.01). Furthermore, we quantify how LLMs infer diagnoses by exploiting retrospective therapeutic cues in clinical narratives, and how temporal misalignment in foundational knowledge bases leads to performance fluctuations. We conclude that prevailing evaluation protocols for augmented LLM systems inadvertently reflect data leakage rather than genuine medical reasoning. To address these systematic artifacts, we propose a rigorous evaluation framework - coupling strict data-isolation protocols with standardized benchmarking safeguards - to ensure future diagnostic AI for rare diseases is evaluated on actual clinical robustness.

arXiv:2610.33544v1 Announce Type: new Abstract: Multi-agent systems have emerged as a promising paradigm for multimodal medical reasoning, but existing methods rely on two assumptions that hinder practical deployment: (i) always-active dense collaboration among multiple agents, and (ii) dependence on large cloud-based models that raise privacy and efficiency concerns. In this paper, we challenge both assumptions and propose SMART-Med, a fully local multi-agent method that treats collaboration as a learnable decision rather than a default behavior. Our key insight is that not every medical query requires costly multi-agent deliberation: many can be reliably solved by a single well-trained agent, while only uncertain cases benefit from expert collaboration. Based on this insight, SMART-Med introduces an uncertainty-driven routing mechanism that adaptively decides when to answer directly and when to recruit a small subset of expert agents, effectively converting dense multi-agent reasoning into selective, query-adaptive coordination. To make this routing reliable, we curate a new multimodal medical dataset MedVR-3K with high-quality reasoning traces and a difficulty hierarchy, and then fine-tune local agents using GRPO with two complementary rewards: a certainty-aware accuracy reward and a vision-language reward model for correctness and reasoning quality respectively. Experiments on four medical visual question answering benchmarks show that SMART-Med achieves competitive performance compared to existing MAS approaches while substantially reducing inference cost (reducing inference time by more than 3.8x and token usage by 72%). Moreover, after fine-tuning on MedVR-3K, SMART-Med yields an average improvement of 6.7% across Slake, PathVQA, VQA-RAD, and PMC-VQA.