7 papers
Continuous-Utility Direct Preference Optimization
Muhammad Ahmed Mohsin, Muhammad Umer, Ahsan Bilal +6
Large language model reasoning is often treated as a monolithic capability, relying on binary preference supervision that fails to capture partial progress or fine-grained reasonin…
: Stratified Scaling Search for Test-Time in Diffusion Language Models
Ahsan Bilal, Muhammad Ahmed Mohsin, Muhammad Umer +6
Test-time scaling investigates whether a fixed diffusion language model (DLM) can generate better outputs when given more inference compute, without additional training. However, n…
Bi-Level Prompt Optimization for Multimodal LLM-as-a-Judge
Bo Pan, Xuan Kan, Kaitai Zhang +6
Large language models (LLMs) have become widely adopted as automated judges for evaluating AI-generated content. Despite their success, aligning LLM-based evaluations with human ju…
On the Fundamental Limits of LLMs at Scale
Muhammad Ahmed Mohsin, Muhammad Umer, Ahsan Bilal +13
Large Language Models (LLMs) have benefited enormously from scaling, yet these gains are bounded by five fundamental limitations: (1) hallucination, (2) context compression, (3) re…
FreeSwim: Revisiting Sliding-Window Attention Mechanisms for Training-Free Ultra-High-Resolution Video Generation
Yunfeng Wu, Jiayi Song, Zhenxiong Tan +2
The quadratic time and memory complexity of the attention mechanism in modern Transformer based video generators makes end-to-end training for ultra high resolution videos prohibit…
Scaling Cross-Embodiment World Models for Dexterous Manipulation
Zihao He, Bo Ai, Tongzhou Mu +6
Cross-embodiment learning seeks to build generalist robots that learn from and operate across diverse morphologies, but differences in kinematics and action spaces hinder data shar…