6 papers
Training Reasoning Models on Saturated Problems via Failure-Prefix Conditioning
Minwu Kim, Safal Shrestha, Anubhav Shrestha +1
As Reinforcement Learning with Verifiable Rewards (RLVR) substantially improves the reasoning abilities of large language models (LLMs), a new bottleneck emerges: more training pro…
On the Limits of Layer Pruning for Generative Reasoning in Large Language Models
Safal Shrestha, Anubhav Shrestha, Aadim Nepal +2
Recent work has shown that layer pruning can effectively compress large language models (LLMs) while retaining strong performance on classification benchmarks, often with little or…
Warm Up Before You Train: Unlocking General Reasoning in Resource-Constrained Settings
Safal Shrestha, Minwu Kim, Aadim Nepal +2
Designing effective reasoning-capable LLMs typically requires training using Reinforcement Learning with Verifiable Rewards (RLVR) or distillation with carefully curated Long Chain…
Layer Importance for Mathematical Reasoning is Forged in Pre-Training and Invariant after Post-Training
Aadim Nepal, Safal Shrestha, Anubhav Shrestha +4
Large language models improve at math after instruction tuning, reinforcement learning, or knowledge distillation. We ask whether these gains come from major changes in the transfo…
Reinforcement Learning vs. Distillation: Understanding Accuracy and Capability in LLM Reasoning
Minwu Kim, Anubhav Shrestha, Safal Shrestha +2
Recent studies have shown that reinforcement learning with verifiable rewards (RLVR) enhances overall accuracy (pass@1) but often fails to improve capability (pass@k) of LLMs in re…
Mathematical Reasoning in Large Language Models: Assessing Logical and Arithmetic Errors across Wide Numerical Ranges
Safal Shrestha, Minwu Kim, Keith Ross
Mathematical reasoning in Large Language Models (LLMs) is often evaluated using benchmarks with limited numerical ranges, failing to reflect real-world problem-solving across diver…