Sección

Investigación

Papers y preprints de los feeds de investigación que seguimos.

6415documentos

Rethinking Agentic Model Orchestration for Time Series Forecasting

Time series foundation models (TSFMs) perform unevenly across forecasting tasks. Recent work uses agentic model orchestration (AMO) to select or combine TSFMs, often outperforming individual forecasters. Yet these gains may reflect stronger or more complementary candidates rather than better orchestration. To disentangle these effects, we introduce FairAMO, a controlled evaluation protocol with frozen candidate forecasts. Experiments on GIFT-Eval show that AMO improves accuracy, but its gains are bounded by the candidate TSFMs. The best orchestrators reduce MASE by less than 6% relative to the best single TSFM across all tasks. These gains depend on complementarity (different candidates excel on different tasks) rather than the number of candidates. Adding members yields diminishing returns, and a redundant member can erase these gains. The value of historical evidence also saturates quickly. A few recent windows capture all gains, and further history adds none. These findings separate system-level forecasting gains from the contribution of orchestration itself.

arXiv AI Papershace 23 días

Parallel Thinking for Large Language Model with Path Diversity

Parallel thinking has emerged as a promising paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs) by exploring multiple reasoning paths simultaneously. However, we identify that parallel thinking via Reinforcement Learning (RL) typically suffers from the mode collapse of reasoning paths, where parallel thinking trajectories degenerate into similar ones. This homogeneity undermines the fundamental motivation of parallel thinking, that is, maintaining multiple hypotheses concurrently to prevent premature loss into a suboptimal solution. In this paper, we propose Diverse Parallel Thinking (DPT), a simple yet effective RL approach for parallel thinking designed to encourage path diversity. DPT rewards the model to develop distinct reasoning strategies by penalizing the prefix similarity of parallel thinking paths, thereby enhancing a more comprehensive exploration of the reasoning space. We conduct various experiments and analyses across multiple reasoning benchmarks to demonstrate the effectiveness of DPT. Results show that DPT successfully outperforms previous parallel thinking methods, achieving state-of-the-art performance over baselines. We validate path diversity as a key element to improve parallel thinking. Our code is available at https://github.com/JiaruQian/DPT.

arXiv AI Papershace 23 días

A Strong MoE Is All You Need for Process Supervision

Process reward models (PRMs) provide fine-grained supervision for multi-step reasoning, yet obtaining reliable process labels remains a central bottleneck. Existing approaches typically rely on human annotation, rollout-based estimation, or auxiliary verification. In this work, we explore a different source of process supervision natively exposed by Mixture-of-Experts (MoE) models: expert routing. We show that simple routing statistics provide an intrinsic process-confidence signal that is competitive with conventional output-logit confidence. Building on this observation, we construct routing confidence by combining overall routing mass, head routing mass, and within-step routing variation, and introduce RoCo-Rethink, which identifies the least confident reasoning step and selectively regenerates the subsequent trajectory. We further introduce RoCo-PRM, which uses a strong MoE to automatically annotate reasoning trajectories with routing-derived step scores and distills them into a standalone PRM. Across multiple MoE architectures and diverse reasoning tasks, both RoCo-Rethink and RoCo-PRM consistently outperform strong baselines, while RoCo-PRM transfers effectively across both model architectures and tasks. Our findings suggest that expert routing is more than a computational dispatch mechanism: it provides an architecture-native source of process supervision for reasoning.

arXiv AI Papershace 23 días

An Exact 93-Move Solution to a Nonlocal Crystal-Exchange Puzzle

We determine the minimum number of directional commands needed to solve alevel of Order of the Sinking Star, modeled as a crystal-exchangepuzzle on an 11*11 board. A command swapsthe player with the nearest crystal in the selected row or column, regardlessof distance; only an empty ray permits an ordinary unit walk. Two gates admitthe player to the exit exactly while all sixteen crystals occupy their targetcells. We give an explicit 93-command solution and exclude every shortersolution by an exact, cost-bounded breadth-first search. The search representsa position by its seventeen occupied cells and one active cell. Its admissibleheuristic partitions command costs between a compressed core abstraction andresidual motion, strengthened by a necessary activation detour and a cornervisit condition. The completed computation retained 605,349,215 occupied setsand 1,863,925,136 states. We prove the reductions and pruning conditions,describe the exact storage scheme, and provide source code, execution logs,and a reproduction driver. The result is a computer-assisted proof for thespecified instance, with an explicit implementation trust boundary.

arXiv AI Papershace 23 días

RubricOpt: Self-Revision of Semantic Rubrics for Preference Optimization

A model's failures can reveal gaps in the behavioral criteria guiding its training. We introduce RubricOpt, a framework that lets a policy revise these criteria within preference optimization. The policy diagnoses its own rubric-free outputs, retains requirements already covered by the rubric, and edits clauses that leave recurring failures unaddressed. A shared semantic representation connects task demands, preference-data contrasts, and rubric clauses, while a deterministic gate checks evidence references and fixed semantic anchors. The rubric then guides Direct Preference Optimization (DPO) during training, with preference pairs, labels, and the objective held fixed, and is removed at evaluation. Our analysis relates the local effect on rubric-free success to semantic alignment and a residual beyond the measured factors. Experiments on capability and open-ended generation benchmarks show that rubric conditioning can improve outcomes over ordinary DPO, with gains and regressions varying across models and datasets. A matched continuation isolates the contribution of revision from additional training and finds mixed incremental effects for the tested edit, whose feedback overlaps the evaluation set. RubricOpt thus brings the behavioral specification itself into the policy's feedback loop.

arXiv AI Papershace 23 días

TenetBench: A Clause-Anchored Russian Corpus for Evaluating Language Models against Professional Codes of Ethics

Codes of professional ethics guide and restrict specialists, who increasingly work with assistants built on large language models. Such an assistant can breach the user's code, or help the user breach it, without the user noticing. Evaluations reporting broad categories of harm cannot show how often this happens or which provision is breached. TenetBench pairs 3,433 requests conflicting with clauses of fifteen Russian-language professional codes with benign twins that differ only in the legitimacy of what is asked. An automated pipeline generates the tasks from the text of a code, so the corpus grows with every code added and a provision yields new tasks on every run. It also produced tasks and human-validated verdicts from two English codes of other jurisdictions. In human validation of the frozen pipeline, specialists rated 0.995 of the conflicting tasks plausible, and the model judge agreed with the adjudicated label at Gwet's AC1 of at least 0.894 on 1,860 elements. We evaluated fourteen model and reasoning-mode combinations on whether they produce the conflicting text, decline or warn. Eleven of the fourteen produce conflicting content on most such requests, mostly without a warning, and nearly three warnings in four accompany conflicting content. The models that conform more do so by declining while still serving the legitimate twin, and enabling reasoning halved the refusals of two models. We hope the dataset and pipeline will help evaluate how language models observe professional norms and let diagnostic datasets be built for particular professions and organisations.

arXiv AI Papershace 23 días

Generative Modeling via Nonlinear Schrödinger Bridges and Branching Diffusions

We propose a new generative modeling framework grounded in forward-backward partial differential equations (PDEs). These equations arise naturally in mean-field game (MFG) theory, and are closely related to nonlinear Schrödinger systems. Our main contribution is to reinterpret this connection within the setting of generative modeling and to develop an efficient numerical methodology for its practical implementation. Specifically, we derive and characterize a nonlinear extension of the classical Schrödinger bridge problem through a coupled system of PDEs, yielding a model we call SBNL (Schrödinger Bridge NonLinear). To compute its solution, we design a simulation algorithm based on branching diffusions and the probabilistic representation of semilinear PDEs. The resulting method is embarrassingly parallel, scalable to high dimensions, and applicable to a broad class of nonlinear interaction functionals. We validate the approach through numerical experiments, including recovery of the classical Schrödinger bridge, nonlinear extensions with convex penalties (Gross-Pitaevskii) as well as potential interaction terms, and high-dimensional generative tasks such as image synthesis on the MNIST and FFHQ human face dataset. Furthermore, the same framework embeds score-based generative modeling, diffusion models, and tractable gaussian Schrödinger bridges.

arXiv AI Papershace 23 días

Density-Level Backward Error for Low-NFE Generative Flows

Low-NFE generative ODE sampling is constrained not only by solver order, but also by how a fixed budget of network evaluations is distributed along the flow. We develop a density-level backward-error framework that identifies the component of a solver's modified-equation defect that changes the evolving marginal, while excluding solenoidal trajectory distortion that is invisible in Wasserstein tangent geometry. The resulting residual induces a solver-order-aware compute-allocation rule and, through gauge freedom, an objective for selecting numerically favorable velocity representatives of the same density path. This separates our approach from schedule search driven by trajectory or endpoint surrogates and from higher-order solver design. Exact Gaussian and mixture models establish the mechanism, including cases where curvature-based allocation fails and continuum-optimal transport is not discretization-optimal. On a pretrained CIFAR-10 EDM probability-flow ODE, density-aware allocation consistently improves fixed-update Euler sampling in the low-NFE regime. Matched-seed convergence to a dense Euler reference confirms that the method accelerates approximation of the same high-NFE flow rather than changing the asymptotic sampler.

arXiv AI Papershace 23 días

Asymmetry Tunneling Beyond Adaptive Equivariance Relaxation

Equivariant neural networks encode prescribed symmetries as powerful inductive biases, yet real-world data may exhibit symmetry-breaking structure that these priors cannot fully capture. Prior work has largely addressed this challenge through model-side equivariance relaxation, seeking an appropriate level of equivariance for the data. This raises a complementary question: must equivariance relaxation be adaptively controlled? To investigate this question, we introduce a residual path alongside a strictly equivariant model and deliberately restrict their interface to examine the transmission of task-relevant structure. Across five benchmarks, we observe both residual collapse and survival despite these restrictions; we call the latter phenomenon asymmetry tunneling. We further characterize this phenomenon through four complementary diagnostics. Our results suggest that adaptive control of equivariance relaxation is not a prerequisite for capturing potential symmetry-breaking structure.

arXiv AI Papershace 23 días

OmniSimulator: Aligning Small Language Models for Authentic Heterogeneous Behavior Modeling

Large language models (LLMs) have shown great potential as general-purpose user simulators for interactive systems. Prior studies have demonstrated that training on high-quality data can substantially improve model reasoning and downstream task performance. However, the upper bound of user simulation is still constrained by the model's reasoning paradigm: native chain-of-thought often degenerates into shallow pattern matching and fails to adequately model the behavioral factors underlying real user decisions. In this paper, we propose OmniSimulator, a novel training framework designed to improve the predictive ability of small LLMs for realistic user simulation on short-video platform environments. Specifically, OmniSimulator organizes behavior-oriented reasoning through a structured schema with five dimensions, carefully selects data from real user-behavior scenarios, and introduces a teacher model to generate high-quality reasoning traces under the proposed schema, helping small models learn behavior-relevant reasoning patterns rather than relying on superficial behavioral correlations. Based on this process, we construct 9,881 high-quality training instances from three months of historical interaction trajectories of 200 real users, averaging about 50 distilled examples per user. Experimental results show that OmniSimulator improves the average score of Qwen3-8B by 34.33%, achieves performance competitive with Claude-Sonnet-4.5, and supports smaller-scale, self-hosted deployment. These results suggest that learning structured behavior-oriented reasoning patterns is a promising direction towards scalable and high-fidelity LLM-based user simulation.

arXiv AI Papershace 23 días

PreCoMem: Predictive Cognitive Memory for Self-Evolving Long-Term Dialogue Agents

Long-horizon dialogue agents need to keep user memories stable enough for personalization while adapting when the user genuinely changes. Existing memory systems often append every observation or overwrite old entries by heuristic rules, which makes transient noise hard to distinguish from real belief drift. We introduce PreCoMem, a memory consolidation framework inspired by predictive coding that represents memory as confidence-weighted beliefs about the user's state. Its core signal, Effective Surprisal, combines semantic deviation and contradiction evidence, then downweights this signal when retrieval is diffuse and unreliable. According to effective surprisal, a two-threshold gate maps each turn to one of three deterministic updates: MAINTAIN reinforces supported beliefs, PROFILE stores ambiguous cues as low-confidence hypotheses, and CORRECT softly decays reliably contradicted facts. The PROFILE stage acts as an explicit waiting room, preventing premature commitment to noisy evidence. Experiments on LoCoMo, LongMemEval-S, and PersonaMem-v2 show consistent state-of-the-art accuracy; our new DiMoBench confirms that PreCoMem adapts to non-stationary belief drift without over-reacting to noise. The code is available at https://github.com/ZhongWishing/PreCoMem.

arXiv AI Papershace 23 días

Harness Scaling for Generalizable Agentic Reinforcement Learning

Agent harnesses have become a central component of modern agentic systems, shaping how models use tools, manage context, and coordinate long-horizon interaction. Realizing the benefits of these harnesses depends on how effectively the underlying model can make use of them, motivating reinforcement learning directly through the harness. However, existing approaches typically train through a single fixed harness, which can lead to specialization and limited transfer across harness configurations. To address this, we introduce Harness Scaling, a new scaling dimension for agentic RL that extends training from a single harness to diverse execution configurations. Concretely, we construct a factorized harness space, select a compact training subset with balanced factorial coverage, and organize cross-harness training to combine global coverage with controlled variation for the same task. Across multiple benchmarks, model scales, and RL algorithms, Harness Scaling consistently improves generalization to unseen white-box configurations. Moreover, models trained through structured white-box variation transfer effectively to unseen external black-box harnesses, outperforming models trained through a single complex black-box harness. These results establish harness diversity as a meaningful scaling dimension for agentic RL and highlight the importance of how this diversity is organized for cross-harness generalization.

arXiv AI Papershace 23 días

RegoMender: A Multi-agent Framework for Automated Rego Policy Repair

Policy-as-Code (PaC) has emerged as a foundational paradigm in cloud-native systems. However, repairing Rego policies remains challenging due to their strict syntax, declarative semantics, and specific verification requirements. Existing automated program repair (APR) methods are mainly designed for mainstream languages and struggle to accommodate the unique policy logic and repair workflows of Rego. Thus, Rego policy repair remains a manual process driven by trial and error. To address this, we introduce RegoBench, the first automated Rego policy benchmark comprising 1,159 real-world defects, and propose RegoMender, a feedback-guided multi-agent framework. Collaborating via a shared blackboard and an atomic toolkit, RegoMender dynamically executes patching, rewriting, or rolling back, to resolve complex repair scenarios. Evaluated on RegoBench, RegoMender achieves syntactic and test-based semantic success rates of 82.9\% and 77.9\%, respectively, outperforming the strongest baseline by 13.4\%. Unlike other approaches whose performance drops in complex scenarios, RegoMender maintains effectiveness across single-step and chained repair tasks. Furthermore, it generalizes strongly across HCL and SQL and achieves an 86.4\% repair rate on 132 real-world CVE vulnerabilities, proving its robustness and practical utility for security.

arXiv AI Papershace 23 días

Omnilingual-MsASR: Benchmarking Multimodal, Multilingual, Long-Form Multi-Speaker ASR of Omni Large Language Models

Omni-modal large language models (LLMs) have achieved remarkable progress in audio and video understanding. However, existing models struggle with long-form, multi-speaker videos, especially in understanding the entire video and recognizing who says what and when. To characterize the capability limit of current omni LLMs and guide future model development, we introduce Omnilingual-MsASR, a multilingual, multimodal benchmark for long-form Multi-speaker Automatic Speech Recognition (MsASR) covering six languages, recordings up to 30 minutes, and scenarios with up to 20 speakers. Omnilingual-MsASR provides parallel audio and video inputs to evaluate models' cross-modal consistency and their ability to exploit complementary cues across modalities. Extensive evaluations of leading omni-modal LLMs reveal substantial performance bottlenecks, while their tcpWERs frequently exceed 100%. These findings highlight fine-grained temporal localization, speaker disambiguation, and cross-modal synergy as key challenges for future omni-modal foundation models. Our work is available at https://github.com/QwenLM/Omnilingua-Bench/tree/main/Omnilingua-MSpeaker.

arXiv AI Papershace 23 días

DynaWeave: Learning to Adapt through Dynamic Scenario Synthesis

Large language model (LLM) agents increasingly use tools to complete tasks in stateful environments. However, many training environments remain largely static apart from changes caused by the agent's own actions, whereas external events introduced by other users, services, or sudden contingencies in real environments can invalidate ongoing plans. To address this gap, we introduce DynaWeave, an automated synthesis framework that transforms existing static tasks into verifiable dynamic scenarios. The framework generates exogenous events relevant to each task, covering changes in task requirements, environment state, and external information. It uses directed acyclic graphs (DAGs) to represent dependencies between events and an execution engine to activate them during agent interaction. A dynamic verifier revises only the evaluation checks affected by events that actually occur and retains all other checks. Using DynaWeave, we collect 2,095 verified interaction trajectories across 36 environments and use them for supervised fine-tuning of Qwen3.5 models. Experiments on Gaia2, CostBench, and ToolMaze show that DynaWeave achieves consistent improvements in overall performance across all three evaluated model sizes, increasing the task completion rate on CostBench by 67.34 percentage points and even exceeding the performance of Gemini-3.1-Pro-Preview on ToolMaze. These results support dynamic scenario synthesis as a practical approach to training agents to adapt to changing environments. Our code is available at https://github.com/Miaow-Lab/DynaWeave.

arXiv AI Papershace 23 días

How Far Can Foundation Models Be Distilled for Efficient Jet Tagging?

Jet tagging is one of the most important applications of machine learning at the Large Hadron Collider. Recent models build in more physics knowledge or rely on large-scale pretraining, but the experiments place strict limits on their inference cost. Knowledge distillation can transfer their performance to small students that are cheap to run. We ask which properties of a teacher help the student, how small a student can get and how close it can come to the state of the art. On a jet tagging benchmark, we distil six teachers that differ in architecture, symmetry and pretraining, including two foundation models, into five small students. Every teacher improves every student, and the gain depends mainly on the strength of the teacher. We then shrink our two strongest distilled students, Transformer and L-GATr-slim, in width and in depth. With 20 thousand parameters, a distilled L-GATr-slim outperforms every published jet tagger trained from scratch at a fraction of the inference cost of every teacher. Even with two thousand parameters, it matches ParT, a model with a thousand times as many parameters. We also study what the students learn from their teachers. Our benchmark of inference cost shows that students of the same size can differ by almost an order of magnitude in speed. We release the code and the weights of all trained models.

arXiv AI Papershace 23 días

Events, Not Pixels: Where World Models Get the Return Wrong

World models are increasingly used for planning, where a planner ranks candidate actions by returns imagined by the model. However, these models are judged mainly by how well they render frames, even though a visually faithful rollout can still assign the wrong return. To identify where this mismatch arises, we introduce a transition-level diagnostic that replaces imagined transitions with real ones and measures the reduction in return error against a budget-matched random control. Our analysis yields two findings. (1) Event transitions, the few steps where a reward is earned or an episode ends, are rendered as well as other steps (median pixel-error ratio 0.85), yet restoring them removes a median of 70% of the return error. In contrast, restoring transitions with the largest MSE, SSIM or LPIPS error removes less than a random choice. Across the 44 Atari-100k model--collection pairs that pass our reliability checks, including three released families, event patches outperform random patches in 39 (sign test p=1.4 x 10^{-7}), with a median gain of +44 percentage points. (2) The remaining error is structured: a complementary patch separates errors at true events from those elsewhere, while an eight-step timing tolerance recovers few events missed by our deterministic models. This structure also guides repair: a rule recorded before four repair outcomes were known predicts when a diffusion model repairs these failures. Event patches improve plan ranking, and the same concentration appears on control tasks with rare reward events.

arXiv AI Papershace 23 días

MERP: A Latent-Bottleneck Framework for Robust Multi-Modal Multi-Label Emotion Recognition

Multimodal multi-label emotion recognition is important for affect-aware systems, but models trained with complete text, visual, and acoustic streams often degrade sharply when some streams are unavailable at deployment. We address this train-test gap by learning a compact representation that preserves emotion-specific evidence from any available subset of modalities. We propose the Multi-modal Emotion Recognition Perceiver (MERP), which maps the joint sequence into a fixed-size latent bank, thereby keeping repeated reasoning independent of raw input length; it extracts label-specific evidence before and after latent reasoning and aligns the two levels through reconstruction. MERP further uses positive and negative class prototypes, a noise-gated feature enhancer with frequency-shaped perturbations during training, and confidence-aware class-wise fusion to strengthen separability, reduce sensitivity to missing inputs, and combine complementary predictors. Under 20 percent test-time random modality drop, MERP achieves micro-F1 scores of 0.563 on CMU-MOSEI and 0.780 on M3ED, exceeding the strongest evaluated baselines by 0.004 and 0.005, respectively; in the severe CMU-MOSEI audio-only setting, it retains 0.473 micro-F1 versus 0.022 for LDDU. These results show that fixed latent reasoning provides a practical, testable route to robust multimodal emotion recognition without requiring missing-modality examples during training. Code is available at https://github.com/Creative-zcx/MERP.

arXiv AI Papershace 23 días

One Act, Many Masks: Effect-Marginal Policy Optimization for Long-Horizon Agents

Recent advances in multi-turn reinforcement learning (RL) have substantially improved the performance of LLM agents on complex, long-horizon interactive tasks. However, despite progress in fine-grained credit assignment, policy optimization still applies these learning signals through response-specific probability ratios. Different textual responses can induce the same executable effect, yet credit differences arising from subsequent outcomes remain tied to their individual textual realizations. We identify this mismatch as credit-update misalignment. Our analysis reveals that credit-text pairing directly shapes policy-update directions, even when the policy context and executable effect are held fixed. Motivated by this finding, we propose Effect-Marginal Policy Optimization (EMPO), which aligns policy optimization with execution-level equivalence through an effect-level surrogate that couples updates across alternative textual realizations. EMPO preserves the original credit signals, integrates with both trajectory-level and step-level credit estimators, and requires no additional critics or rollouts. Experiments on ALFWorld, WebShop, and Sokoban show that EMPO outperforms strong baselines in both task performance and learning efficiency.

arXiv AI Papershace 23 días

BIRD-RL: Scaling Agentic Reinforcement Learning over Stateful Data-Centric Environments

While agentic reinforcement learning (RL) has emerged as a promising paradigm for training small language models (SLMs), existing frameworks are predominantly tool-centric and struggle with data-centric environments. Specifically, applying tool-centric frameworks to stateful tasks like SQL debugging introduces severe spatial inefficiencies since proper environment isolation typically requires a separate database container per rollout. Naively reusing databases without tracking evolving states across multi-turn interactions causes trajectory contamination and cross-trajectory interference, which severely destabilizes policy updates. To overcome these bottlenecks, we introduce BIRD-RL, a novel four-stage agentic RL curriculum that alternately enhances foundational agent reasoning and trajectory-adaptive on-policy exploration to circumvent the instability inherently associated with direct end-to-end learning. Crucially, to support this curriculum, BIRD-RL is driven by a Trajectory-scoped Persistent Execution Infrastructure, which enables safe database replica reuse while preserving trajectory isolation. This design improves spatial RL efficiency by 12.8x, successfully enabling scalable agentic RL for complex data-centric tasks without needs of Kubernettes. Empirical evaluations on the BIRD-CRITIC-SQLite benchmark demonstrate that our 7B and 14B models achieve Success Rates of 44.40% and 48.00%, respectively, showing competitive performance against strong proprietary model-based agents such as Claude-Opus-4.6 Agent (48.20%). Furthermore, BIRD-RL successfully trains unified SLMs capable of both advanced SQL debugging and generation, highlighting its broader potential for multi-objective optimization across heterogeneous, real-world data-centric tasks.