publications

2025

  1. ACL
    ELI-Why: Evaluating the Pedagogical Utility of LLM Explanations
    Brihi Joshi*, Keyu He*, Sahana Ramnath, Sadra Sabouri, Kaitlyn Zhou, Souti Chattopadhyay, Swabha Swayamdipta, and Xiang Ren
    Findings of ACL, 2025
  2. PREPRINT
    Amulet: Putting Complex Multi-Turn Conversations on the Stand with LLM Juries
    Sahana Ramnath, Anurag Mudgil, Brihi Joshi, Skyler Hallinan, and Xiang Ren
    Under Review, 2025
  3. PREPRINT
    Believing without Seeing: Quality Scores for Contextualizing Vision-Language Model Explanations
    Keyu He, Tejas Srinivasan, Brihi Joshi, Xiang Ren, Jesse Thomason, and Swabha Swayamdipta
    Under Review, 2025
  4. PREPRINT
    Improving LLM Personas via Rationalization with Psychological Scaffolds
    Brihi Joshi, Xiang Ren, Swabha Swayamdipta, Rik Koncel-Kedziorski, and Tim Paek
    Under Review, 2025
  5. PREPRINT
    PrimeX: A dataset of worldview, opinion, and explanation
    Rik Koncel-Kedziorski, Brihi Joshi, and Tim Paek
    Under Review, 2025
  6. PREPRINT
    Position: Stop Using Brittle Interpretations
    Isabelle Lee, Emmy Liu, Cathy Jiao, Michael Saxon, Brihi Joshi, Dani Yogatama, Fazl Barez, and Naomi Saphra
    Under Review, 2025

2024

  1. EMNLP
    OATH-Frames: Characterizing Online Attitudes towards Homelessness via LLM Assistants
    Jaspreet Ranjit, Brihi Joshi, Rebecca Dorn, Laura Petry, Olga Koumoundouros, Jayne Bottarini, Peichen Liu, Eric Rice, and Swabha Swayamdipta
    EMNLP, Outstanding Paper Award, 2024
    HTML
  2. ICLR
    Tailoring Self-Rationalizers with Multi-Reward Distillation
    Sahana Ramnath, Brihi Joshi, Skyler Hallinan, Ximing Lu, Liunian Harold Li, Aaron Chan, Jack Hessel, Yejin Choi, and Xiang Ren
    ICLR, 2024
    HTML

2023

  1. ACL
    XMD: An End-to-End Framework for Interactive Explanation-Based Debugging of NLP Models
    Dong-Ho Lee*, Akshen Kadakia*, Brihi Joshi, Aaron Chan, Ziyi Liu, Kiran Narahari, Takashi Shibuya, Ryosuke Mitani, Toshiyuki Sekiya, Jay Pujara, and Xiang Ren
    ACL Demo Track, 2023
    HTML
  2. ACL
    Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-text Rationales
    Brihi Joshi*, Ziyi Liu*, Sahana Ramnath, Aaron Chan, Zhewei Tong, Shaoliang Nie, Qifan Wang, Yejin Choi, and Xiang Ren
    ACL and trAIt@CHI, 2023
    HTML
  3. PREPRINT
    KNIFE: Knowledge Distillation with Free-Text Rationales
    Aaron Chan*, Zhiyuan Zeng*, Wyatt Lake, Brihi Joshi, Hanjie Chen, and Xiang Ren
    TrustML-(un)Limited@ICLR, 2023
    HTML

2022

  1. EMNLP
    ER-Test: Evaluating Explanation Regularization Methods for NLP Models
    Brihi Joshi*, Aaron Chan*, Ziyi Liu*, Shaoliang Nie, Maziar Sanjabi, Hamed Firooz, and Xiang Ren
    Findings of EMNLP and TrustNLP@NAACL, 2022
    HTML

2020

  1. COLING
    The Devil is in the Details: Evaluating Limitations of Transformer-based Methods for Granular Tasks
    Brihi Joshi, Neil Shah, Francesco Barbieri, and Leonardo Neves
    International Conference on Computational Linguistics, 2020
    HTML Code
  2. ACM TIST
    Analyzing and Detecting Collusive Users Involved in Blackmarket Retweeting Activities
    Udit Arora, Hridoy Sankar Dutta, Brihi Joshi*, Aditya Chetan*, and Tanmoy Chakraborty
    ACM Transactions on Intelligent Systems and Technology, 2020
    HTML Code
  3. ISMIR
    Did You “Read” the Next Episode? Using Textual Cues for Predicting Podcast Popularity
    Brihi Joshi, Shravika Mittal, and Aditya Chetan
    Workshop on NLP for Music and Audio (NLP4MusA), ISMIR, 2020
    HTML Code
  4. ICMI
    It’s Not What They Play, It’s What You Hear: Understanding Perceived vs. Induced Emotions in Hindustani Classical Music
    Amogh Gulati, Brihi Joshi, Chirag Jain, and Jainendra Shukla
    International Conference on Multimodal Interaction Late Breaking Reports, 2020
    HTML

2019

  1. WSDM
    CoReRank: Ranking to Detect Users Involved in Blackmarket-Based Collusive Retweeting Activities
    Aditya Chetan*, Brihi Joshi*, Hridoy Sankar Dutta*, and Tanmoy Chakraborty
    International Conference on Web Search and Data Mining, 2019
    HTML Code

2018

  1. ASONAM
    Retweet Us, We will Retweet You: Spotting Collusive Retweeters Involved in Blackmarket Services
    Hridoy Sankar Dutta, Aditya Chetan, Brihi Joshi, and Tanmoy Chakraborty
    International Conference on Advances in Social Networks Analysis and Mining, 2018
    HTML Code
  2. PREPRINT
    Generating Clues for Gender based Occupation De-biasing in Text
    Nishtha Madaan, Gautam Singh, Sameep Mehta, Aditya Chetan, and Brihi Joshi
    arXiv:1804.03839, 2018
    HTML