About
Research & Development
Evaluating Long-Context Language Models with AcademicEval: A New Live Benchmark
TLDR: AcademicEval is a new live benchmark for evaluating Large Language Models (LLMs) on long-context academic writing tasks (Title, Abstract, Introduction, Related Work). It...
Research & Development
SoftMimic: Enabling Humanoid Robots to Interact Gently and Safely
TLDR: SoftMimic is a new framework that teaches humanoid robots to respond compliantly to external forces while performing tasks. Unlike traditional stiff controllers, SoftMimic...
Research & Development
Enhancing Large Vision-Language Model Safety with Prompt-Guided Tree Search
TLDR: VisuoAlign is a novel framework that significantly improves the safety of Large Vision-Language Models (LVLMs) by integrating safety constraints directly into their reasoning...
Research & Development
LLMs Face Reasoning Limits Even with Interactive Environments
TLDR: A study on Large Language Models (LLMs) using the Tower of Hanoi puzzle found that providing an interactive environment does not prevent performance...
Research & Development
PROOFFLOW: A Dependency Graph Approach to Faithful Proof Autoformalization
TLDR: PROOFFLOW is a novel pipeline that translates natural language mathematical theorems and proofs into machine-verifiable code, prioritizing the preservation of the original proof's...
Research & Development
Keeping Digital Replicas on Track: A New Approach for Simulating Complex Systems
TLDR: PAINT (Parallel-in-time Neural Twins) is a novel method for creating accurate digital replicas of real-world systems, called Neural Twins. Unlike traditional models that...