Reinforcement Learning with Segment Reward Feedback under Linear Function Approximation
Research
Research News
New papers, breakthroughs and academic AI research
Rethinking Visual Provenance: Detection and Watermarking Across Direct Visual Generation and LLM-Driven Code Rendering
Research
Cross-Lingual Activation Steering for Multilingual Language Models
Research
Unbiased Reward Modeling from Implicit Feedback for LLM Alignment
Research
Attenuated in-context identification in time-series foundation models: diagnosis under counterfactual inputs and repair by synthetic forced-system fine-tuning
Research
Real-Time Generation of Game Video Commentary with Multimodal LLMs: Pause-Aware Decoding Approaches
Research
Performative Prediction with Selective Labels
Research
Quantifying the Gap between Understanding and Generation within Unified Multimodal Models
Research
On the Intrinsic Limited Robustness of Latent-Based Watermarking
Research
Model in Distress: Sentiment Analysis on French Synthetic Social Media
Research
Local Content-Style Control for Diffusion-based Image Stylization
Research
Finding the Heads and the Neurons Responsible for Network Information Retrieval in Language Models
Research