GPQA-D
PulseAugur coverage of GPQA-D — every cluster mentioning GPQA-D across labs, papers, and developer communities, ranked by signal.
-
New TRAAC method optimizes LLM reasoning by compressing thought processes
Researchers have developed a new post-training reinforcement learning method called TRAAC (Think Right with Adaptive, Attentive Compression) to address under- and overthinking in large language models. TRAAC uses self-a…
-
New research tackles LLM agent challenges in edge-cloud, robustness, and reliability
Recent research explores various aspects of Large Language Model (LLM) agents, focusing on improving their performance, robustness, and reliability. Studies introduce benchmarks like EC2-Bench for edge-cloud collaborati…
-
AI benchmark scores predictable from just two factors, study finds
A new research paper proposes a method called BenchPress that can predict a frontier model's performance across numerous benchmarks using only two key scores. The study analyzed 84 models and 133 benchmarks, finding tha…