Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2605. 21751v2 Announce Type: replace Abstract: Text-to-optimization requires two separable capabilities: modeling -- choosing the right optimization structure -- and binding -- grounding every coefficient, index, and parameter in the concrete problem data.
arXiv:2607. 15200v1 Announce Type: cross Abstract: Reinforcement learning has proven effective for improving reasoning in large language models, but extending it to Masked Diffusion Language Models (MDLMs) remains challenging due to the intractability of the log-likelihood estimation.
arXiv:2607. 15142v1 Announce Type: new Abstract: World models are usually evaluated as components of model-based reinforcement learning (MBRL) systems, while the world models themselves are rarely studied in isolation.
arXiv:2607. 14145v1 Announce Type: new Abstract: Tool-augmented large language model agents excel at long-horizon tasks, yet they are typically post-trained on fixed toolsets.
arXiv:2605. 25170v2 Announce Type: replace-cross Abstract: Training data for olfaction is scattered through disparate, non-standardized datasets that limit the ability to build representative world models.
arXiv:2406. 04737v2 Announce Type: cross Abstract: The cellular network of magnetic Induction (MI) communication holds promise in long-distance underground environments.
arXiv:2607. 14886v1 Announce Type: new Abstract: Temporal Knowledge Graph (TKG) reasoning under the extrapolation setting focuses on forecasting future time-stamped events (facts) from historical data in a temporal knowledge graph.
arXiv:2505. 22442v3 Announce Type: replace-cross Abstract: Offline RL (ORL) promises safe and sample-efficient deployment but existing methods rely on undocumented online interactions for hyperparameter tuning and lack reliable fully offline estimates of initial online performance.
arXiv:2607. 14614v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) commonly uses entropy for advantage shaping.
arXiv:2607. 15273v1 Announce Type: cross Abstract: MeanFlow generators achieve fast few-step sampling by predicting average velocities over time intervals, making them attractive for efficient generation.
arXiv:2607. 14506v1 Announce Type: cross Abstract: While reinforcement learning with verifiable rewards (RLVR) is widely used to improve the reasoning capabilities of large language models (LLMs), the generalizability of the resulting models remains poorly understood.
arXiv:2607. 14962v1 Announce Type: cross Abstract: Text-to-image (T2I) models can synthesize realistic, prompt-aligned images, yet samples generated for the same prompt often cover only a small subset of visually distinct modes.
arXiv:2607. 15003v1 Announce Type: new Abstract: The deployment of autonomous cyber-physical systems in safety-critical environments requires closed-loop control strategies (i.
arXiv:2607. 14185v1 Announce Type: cross Abstract: Feedback-driven loops support iterative improvement in large language models, reinforcement learning, and autonomous discovery, yet their gains often diminish under repeated internal feedback.
arXiv:2607. 14192v1 Announce Type: new Abstract: As recommender systems mature in the past few years, their optimization objectives have evolved from a primary focusing on short-term behavioral signals to a broader emphasis on long-term user engagement and retention.
arXiv:2607. 15107v1 Announce Type: new Abstract: This paper develops a categorical framework -- Learning in Infinitesimal Non-Compositional Sketches (LINCS) -- as the repair of non-compositionality: failures of diagrams to factor through quotient sketches lifted to the tangent category setting.
arXiv:2607. 14272v1 Announce Type: new Abstract: Flow matching has emerged as an effective framework for learning complex data distributions, but adapting pretrained flow models to new tasks often requires computationally expensive retraining.
arXiv:2607. 13501v2 Announce Type: replace Abstract: Reinforcement learning for multi-turn search reasoning typically relies on terminal outcome rewards, which cannot distinguish useful, redundant, and harmful intermediate interactions.
arXiv:2607. 14682v1 Announce Type: new Abstract: Efficient multimodal document question answering with explicit visual grounding, locating the precise document region that supports each answer remains an open challenge.
arXiv:2604. 09870v2 Announce Type: replace-cross Abstract: We investigate how looped transformers encode human preference, training lightweight evaluator heads on frozen Ouro-2.