Post-training algorithms based on deep reinforcement learning can push the limits of robotic models for specific objectives, such as generalizability, accuracy, and robustness. However, Intervention-requiring Failures (IR Failures) (e.g., a robot spilling water or breaking fragile glass) during real-world exploration happen inevitably, hindering the practical deployment of such a paradigm. To tackle this, we introduce Failure-Aware Offline-to-Online Reinforcement Learning (FARL), a new paradigm minimizing failures during real-world reinforcement learning. We create FailureBench, a benchmark that incorporates common failure […]

Ver mais

arXiv:2601.08839v1 Announce Type: new Abstract: This paper presents a tri-agent cross-validation framework for analyzing stability and explainability in multi-model large language systems. The architecture integrates three heterogeneous LLMs-used for semantic generation, analytical consistency checking, and transparency auditing-into a recursive interaction cycle. This design induces Recursive Knowledge Synthesis (RKS), where intermediate representations are continuously refined through mutually constraining transformations irreducible to single-model behavior. Across 47 controlled trials using public-access LLM deployments (October 2025), we evaluated system stability via four […]

Ver mais

arXiv:2502.09986v4 Announce Type: replace-cross Abstract: Getting tools that allow simple representations and comparisons of a set of categorical trajectories is of major interest for statisticians. Without loosing any information, we associate to each state a binary random indicator function, taking values in ${0,1}$, and turn the problem of statistical description of the categorical trajectories into a multivariate functional principal components analysis. This viewpoint encompasses experimental frameworks where two or more states can be observed simultaneously. The sample paths […]

Ver mais

We demonstrate deterministic localization of cryptographic hash preimages within specific layers of deep neural networks trained on information-geometric principles. Using a modified Spin-Glass architecture, MD5 and SHA-256 password preimages are consistently identified in layers ES15-ES20 with >90% accuracy for passwords and >85% for hash values. Analysis reveals linear scaling where longer passwords occupy proportionally expanded layer space, with systematic replication in higher-dimensional layers showing exact topological correspondence.Critically, independent network runs with fresh initialization maintain 41.8% information persistence across […]

Ver mais

arXiv:2602.11215v1 Announce Type: new Abstract: While large language models (LLMs) have achieved strong performance through fine-tuning within individual scientific domains, their learning dynamics in multi-disciplinary contexts remains poorly understood, despite the promise of improved generalization and broader applicability through cross-domain knowledge synergy. In this work, we present the first systematic study of multi-disciplinary LLM fine-tuning, constructing a five-discipline corpus and analyzing learning patterns of full fine-tuning, LoRA, LoRA-MoE, and LoRA compositions. Particularly, our study shows that multi-disciplinary learning […]

Ver mais

Reinforcement Learning with Verifiable Rewards (RLVR) has improved the reasoning abilities of large language models (LLMs) on mathematics and programming tasks, but standard approaches that optimize single-attempt accuracy can inadvertently suppress response diversity across repeated attempts, narrowing exploration and overlooking underrepresented strategies. We introduce UpSkill, a training time method that adapts Mutual Information Skill Learning (MISL) to LLMs for optimizing pass@k correctness. We propose a novel reward that we implement within Group Relative Policy Optimization (GRPO): a token-level […]

Ver mais

arXiv:2601.03261v1 Announce Type: new Abstract: Deep Research agents predominantly optimize search policies to maximize retrieval probability. However, we identify a critical bottleneck: the retrieval-utilization gap, where models fail to use gold evidence even after it is retrieved, due to context blindness in noisy environments. To bridge this gap, we propose DeepResearch-Slice, a simple yet effective neuro-symbolic framework. Unlike implicit attention, our approach predicts precise span indices to perform a deterministic hard filter before reasoning. Extensive evaluations across six […]

Ver mais

Preference-Conditioned Policy Learning (PCPL) in Multi-Objective Reinforcement Learning (MORL) aims to approximate diverse Pareto-optimal solutions by conditioning policies on user-specified preferences over objectives. This enables a single model to flexibly adapt to arbitrary trade-offs at run-time by producing a policy on or near the Pareto front. However, existing benchmarks for PCPL are largely restricted to toy tasks and fixed environments, limiting their realism and scalability. To address this gap, we introduce GraphAllocBench, a flexible benchmark built on a […]

Ver mais