…Xuanfei Ren , Abstract Offline reinforcement learning with trajectory-level outcome supervision presents statistical challenges that can be addressed through pessimistic actor-critic methods, though fundamental barriers exist for certain generalized outcome-based…
…We also design the first agentic streaming inference framework that supports thousand-second-scale or even longer generation while mitigating drift with agentic cache management and prompt planing. These innovations significantly accelerate…
…COIL-20 datasets with additional transformation dimensions and dense sampling, enabling accurate estimation of curvature, reach, and volume for theoretical analysis and validation. Generated by Qwen/Qwen2.5-Coder-32B-Instruct A…
…Coverage And Precision Evaluation), with human-written ground-truth captions and human-verified atomic checklist items. Specifically, CAPEval decomposes caption quality into Coverage and Precision . The former quantifies how thoroughly a caption…
…aligned adversarial learning, step-decoupled parameterization, and end-to-end training with iterative regularization. Generated by Qwen/Qwen2.5-Coder-32B-Instruct Few-step diffusion distillation has become increasingly mature for 4…
…Generated by Qwen/Qwen2.5-Coder-32B-Instruct Large Language Model (LLM) safety has often been evaluated at the behavior level, which provides limited evidence of internal robustness, as these evaluations target…
…Strong open reasoning models combined with next-gen creative tools—like Experience the future of AI image editing —open up some very compelling hybrid workflows for agents, content pipelines, and local-first…
…compression method combines pruning with structural minimization to create significantly smaller neural networks while maintaining accuracy, extending to transformer architectures through CompensatedLayerNorm. Generated by Qwen/Qwen2.5-Coder-32B-Instruct Unstructured pruning…
…Evaluating Demographic Perspective-Taking in LLM Hate Speech Annotation (2026) Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification (2026) The Silent Vote: Improving Zero-Shot LLM Reliability by Aggregating Semantic…
…reveals inconsistent documentation of critical annotation details, with improvements over time but ongoing gaps in reproducibility and reliability. Generated by Qwen/Qwen2.5-Coder-32B-Instruct Human annotation is the empirical foundation…