Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
RoPE-LIME: RoPE-Space Locality + Sparse-K Sampling for Efficient LLM Attribution
Isaac Picov, Ritesh Goru
Explaining closed-source Large Language Model (LLM) outputs is challenging because API access prevents gradient-based attribution, while perturbation methods are costly and noisy w…
cs.CL2025
One-Pass to Reason: Token Duplication and Block-Sparse Mask for Efficient Fine-Tuning on Multi-Turn Reasoning
Ritesh Goru, Shanay Mehta, Prateek Jain
Fine-tuning Large Language Models (LLMs) on multi-turn reasoning datasets requires N (number of turns) separate forward passes per conversation due to reasoning token visibility co…