…The following papers were recommended by the Semantic Scholar API Deliberative Alignment is Deep, but Uncertainty Remains: Inference time safety improvement in reasoning via attribution of unsafe behavior to base model (2026…
…Existing failure detection methods either rely on expensive action resampling or external models, while alternatives propagate trajectory-level labels uniformly across every timestep, obscuring localized failure signals. In this paper, we propose…
…Generated by Qwen/Qwen2.5-Coder-32B-Instruct Training accurate medical image segmentation models requires large amounts of densely annotated data, which is costly and time-consuming to obtain. Semi-supervised learning…
…At each window expansion step, the method integrates local uncertainty with structural signals via a unified mechanism that supports dynamic structured generation, including both flexible block expansion and block organization, while maintaining…
…AI-generated summary Recent advances in generative video models are increasingly driven by post-training and test-time scaling, both of which critically depend on the quality of video reward models (RMs…
…The following papers were recommended by the Semantic Scholar API Adaptive Test-Time Compute Allocation for Reasoning LLMs via Constrained Policy Optimization (2026) Uncertainty-Aware Budget Allocation for Adaptive Test-Time Reasoning…
…Generated by Qwen/Qwen2.5-Coder-32B-Instruct Selecting the best response from multiple small-model samples using a stronger scorer is a simple inference-time strategy, but fails when the small…
…it adds only 5.2% to the captioning time, compared with 65.4% for CSTA and 211.9% for MaxInfo. We release our code and pre-trained checkpoint at https://github.com…
…This motivates treating teacher exposure not as a fixed hyperparameter but as a learnable training-time control variable. We therefore propose Adaptive Teacher Exposure for Self-Distillation (ATESD). ATESD models the reveal…
…Existing scalar, score-token, and pairwise reward models over-compress uncertainty and fine-grained score differences, while reasoning-based generative rewards provide stronger judgments but are costly to deploy and difficult to…