activity
20202026
most citedThe GEM Benchmark: Natural Language Generation, its Evaluation and Metrics

52 citations · 89 across the 26 of their papers we have counts for

collaborators
Showing cs.CLShow all

13 papers · 1 filter

cs.CL2026

Last Translation Benchmark

Vilém Zouhar, Niyati Bafna, Mukund Choudhary +241

For scientific progress, we need benchmarks that test the limits of state-of-the-art models, and evaluation methods that inform us about failure cases. As models get stronger, stan…

cs.CL2026

How Fine-Grained Should a RAG Benchmark Be? A Hierarchical Framework for Synthetic Question Generation

Chase M. Fensore, Kaustubh Dhole, Jason Fan +2

Evaluating retrieval-augmented generation (RAG) systems requires benchmarks that capture diverse question characteristics, yet practitioners lack empirical guidance on which dimens…

cs.CL2026

BabyReasoningBench: Generating Developmentally-Inspired Reasoning Tasks for Evaluating Baby Language Models

Kaustubh D. Dhole

Traditional evaluations of reasoning capabilities of language models are dominated by adult-centric benchmarks that presuppose broad world knowledge, complex instruction following,…

cs.CL2025

Adversarial Lens: Exploiting Attention Layers to Generate Adversarial Examples for Evaluation

Kaustubh Dhole

Recent advances in mechanistic interpretability suggest that intermediate attention layers encode token-level hypotheses that are iteratively refined toward the final output. In th…

cs.CL2025

Toward Honest Language Models for Deductive Reasoning

Jiarui Liu, Kaustubh Dhole, Yingheng Wang +7

Deductive reasoning is the process of deriving conclusions strictly from the given premises, without relying on external knowledge. We define honesty in this setting as a model's a…

cs.CL2025

Evaluating Hybrid Retrieval Augmented Generation using Dynamic Test Sets: LiveRAG Challenge

Chase Fensore, Kaustubh Dhole, Joyce C Ho +1

We present our submission to the LiveRAG Challenge 2025, which evaluates retrieval-augmented generation (RAG) systems on dynamic test sets using the FineWeb-10BT corpus. Our final…