Benjamin Van Durme
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Benjamin Van Durme has a total of 124 co-authors. Here are the top co-authors by number of shared publications:
- Kate Sanders
Johns Hopkins University
7 shared publications Seth Ebner
Johns Hopkins University
7 shared publications- Nathaniel Weir
Johns Hopkins University
6 shared publications Yunmo Chen
Johns Hopkins University
5 shared publications- Elias Stengel-Eskin
Johns Hopkins University Johns Hopkins University
4 shared publications - Eugene Yang
Johns Hopkins University, Baltimore, MD, USA
4 shared publications
Latest Publications
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 10543-10554
ARIA: A Causal-Aware Framework for Rescuing LLM Reasoning in Trustworthy Materials Discovery
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 10543-10554
Open AccessARIA: A Causal-Aware Framework for Rescuing LLM Reasoning in Trustworthy Materials Discovery
- Open Access
- 24 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 337-348
Beyond Relevance: On the Relationship Between Retrieval and RAG Information Coverage
- 24 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 337-348
Open AccessBeyond Relevance: On the Relationship Between Retrieval and RAG Information Coverage
- Open Access
- 24 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 151-158
A Replicability Study of XTR
- 24 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 151-158
Open AccessA Replicability Study of XTR
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1519-1530
Multi-Vector Index Compression in Any Modality
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1519-1530
Open AccessMulti-Vector Index Compression in Any Modality
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3388-3395
CoverageBench: Evaluating Information Coverage across Tasks and Domains
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3388-3395
Open AccessCoverageBench: Evaluating Information Coverage across Tasks and Domains
- 24 March 2026
- Published by Springer Nature
- p. 305-318
Does Reasoning Make Search More Fair? Comparing Fairness in Reasoning and Non-reasoning Rerankers
- 24 March 2026
- Published by Springer Nature
- p. 305-318
Does Reasoning Make Search More Fair? Comparing Fairness in Reasoning and Non-reasoning Rerankers
- Open Access
SocialNLI: A Dialogue-Centric Social Inference Dataset
Making theory-of-mind inferences from human dialogue is a strong indicator of a model's underlying social abilities, which are fundamental for adept AI assistants. However, large language and reasoning models struggle to understand sophisticated social phenomena in transcript data, such as sarcasm and irony. To assess the weaknesses of current models and to identify their solutions, we introduce SocialNLI (SoNLI) -- the first social dialogue inference dataset. SoNLI consists of a collection of dialogue transcripts hand-picked to center complex social nuances like irony and sarcasm, paired with inferences, corresponding likelihood scores, and human-written explanations. We explore social inference analysis as a facet of theory-of-mind, and evaluate LLM and reasoning model theory-of-mind ability through multi-step counterfactual reasoning.
Open AccessSocialNLI: A Dialogue-Centric Social Inference Dataset
Making theory-of-mind inferences from human dialogue is a strong indicator of a model's underlying social abilities, which are fundamental for adept AI assistants. However, large language and reasoning models struggle to understand sophisticated social phenomena in transcript data, such as sarcasm and irony. To assess the weaknesses of current models and to identify their solutions, we introduce SocialNLI (SoNLI) -- the first social dialogue inference dataset. SoNLI consists of a collection of dialogue transcripts hand-picked to center complex social nuances like irony and sarcasm, paired with inferences, corresponding likelihood scores, and human-written explanations. We explore social inference analysis as a facet of theory-of-mind, and evaluate LLM and reasoning model theory-of-mind ability through multi-step counterfactual reasoning.
- Open Access
MegaWika 2: A More Comprehensive Multilingual Collection of Articles and their Sources
We introduce MegaWika 2, a large, multilingual dataset of Wikipedia articles with their citations and scraped web sources; articles are represented in a rich data structure, and scraped source texts are stored inline with precise character offsets of their citations in the article text. MegaWika 2 is a major upgrade from the original MegaWika, spanning six times as many articles and twice as many fully scraped citations. Both MegaWika and MegaWika 2 support report generation research ; whereas MegaWika also focused on supporting question answering and retrieval applications, MegaWika 2 is designed to support fact checking and analyses across time and language.
Open AccessMegaWika 2: A More Comprehensive Multilingual Collection of Articles and their Sources
We introduce MegaWika 2, a large, multilingual dataset of Wikipedia articles with their citations and scraped web sources; articles are represented in a rich data structure, and scraped source texts are stored inline with precise character offsets of their citations in the article text. MegaWika 2 is a major upgrade from the original MegaWika, spanning six times as many articles and twice as many fully scraped citations. Both MegaWika and MegaWika 2 support report generation research ; whereas MegaWika also focused on supporting question answering and retrieval applications, MegaWika 2 is designed to support fact checking and analyses across time and language.
- Open Access
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2632-2636
RE-AdaptIR: Improving Information Retrieval through Reverse Engineered Adaptation
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2632-2636
Open AccessRE-AdaptIR: Improving Information Retrieval through Reverse Engineered Adaptation
- Open Access
- 16 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 425-429
LLMs Provide Unstable Answers to Legal Questions
- 16 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 425-429
Open AccessLLMs Provide Unstable Answers to Legal Questions