5 papers
PhantomEnvironments: Training LLM Agents in Fictional Worlds
Anmol Kabra, Swathi Saravana Selvam, Albert Gong +5
Training LLM agents with reinforcement learning (RL) is bottlenecked by environments, which must provide verifiable rewards, support long-horizon interaction, and scale cheaply. Ex…
Learning from Synthetic Data Improves Multi-hop Reasoning
Anmol Kabra, Yilun Yin, Albert Gong +6
Reinforcement Learning (RL) has been shown to significantly boost reasoning capabilities of large language models (LLMs) in math, coding, and multi-hop reasoning tasks. However, RL…
Electron-affinity difference distributions as an organizing principle for superconductivity, enabling the discovery of PtPbBi
Omri Lesser, Yanjun Liu, Natalie Maus +11
Predicting the superconducting transition temperature () from crystal structure and composition remains a central challenge in condensed-matter physics, reflecting the absence…
PhantomWiki: On-Demand Datasets for Reasoning and Retrieval Evaluation
Albert Gong, Kamilė Stankevičiūtė, Chao Wan +6
High-quality benchmarks are essential for evaluating reasoning and retrieval capabilities of large language models (LLMs). However, curating datasets for this purpose is not a perm…
Domain Private Transformers for Multi-Domain Dialog Systems
Anmol Kabra, Ethan R. Elenberg
Large, general purpose language models have demonstrated impressive performance across many different conversational domains. While multi-domain language models achieve low overall…