collaborators

5 papers

cs.LG2026

Quality-Diversity Stress Tests for Process Reward Models:What Archive Coverage Can and Cannot Certify

Ibne Farabi Shihab, Fariya Afrin

Process reward models (PRMs) score intermediate reasoning steps and are widely used for search, ranking, and training, but optimization can exploit these learned proxies by increas…

cs.LG2026

Evaluator Ensembles Under Reward Hacking: Covariance Geometry and Finite-Search Guarantees

Fariya Afrin, Ibne Farabi Shihab

Language-model judges and reward models enable scalable supervision, but finite optimization can exploit evaluator errors rather than improve response quality. We characterize this…

cs.LG2026

EST-PRM: Stress-Testing Process Reward Models Before They Become Load-Bearing

Ibne Farabi Shihab, Fariya Afrin, Sanjeda Akter +1

Process reward models (PRMs) are widely used in language-model training with dense step-level supervision. They assume PRM scores are stable proxies for step correctness under labe…

cs.LG2026

Grounded Decoding: Retrieval-Anchored Probability Fusion for Faithful RAG

Ibne Farabi Shihab, Fariya Afrin, Sanjeda Akter +1

As retrieval-augmented generation (RAG) systems scale, it becomes increasingly challenging to ensure faithful grounding in external evidence. Large language models may still priori…

cs.LG2026

Dynamic Proxy-Mixing: Transferring Replay Controllers from Small to Large Models for Continual Instruction Tuning

Ibne Farabi Shihab, Fariya Afrin, Anuj Sharma

Continual instruction tuning updates a language model through a sequence of new domains, yet each update can progressively erode previously learned capabilities and alignment behav…