Unlocking Lossless Speedups in LLMs via Discrete Diffusion Paper • 2609.04010 • Published 7 days ago • 135
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning Paper • 2607.02509 • Published Jul 2 • 2
PAPERMIND: Benchmarking Agentic Reasoning and Critique over Scientific Papers in Multimodal LLMs Paper • 2604.21304 • Published Apr 23
Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders Paper • 2410.20526 • Published Oct 27, 2024 • 1