Samyadeep Basu

A Survey on LLM-based Conversational User Simulation

Bo Ni, Leyao Wang, Yu Wang, Branislav Kveton et autres

User simulation has long played a vital role in computer science due to its potential to support a wide range of applications. Language, as the primary medium of human communication, forms the foundation of social interaction and behavior. Consequently, simulating conversational behavior …

Sparse Personalized Text Generation with Multi-Trajectory Reasoning

Bo Ni, Haowei Fu, Qinwen Ge, Franck Dernoncourt et autres

As Large Language Models (LLMs) advance, personalization has become a key mechanism for tailoring outputs to individual user needs. However, most existing methods rely heavily on dense interaction histories, making them ineffective in cold-start scenarios where such data is sparse or unavailable. …

Figure Captioning with Large Generative Models: A Survey

Chan-Wei Hu, Reuben Luera, Franck Dernoncourt, Huanjiang Liu et autres

Figures, such as charts, graphs, and scientific visualizations, are essential for communicating data-driven insights, yet their interpretation often depends on clear and informative captions. Recent advances in large generative models, particularly multimodal large language models (MLLMs), have enabled significant progress in automatic …

us, hk, kr, ca, se, mx (code pays fourni par la source)

A Survey on LLM-based Conversational User Simulation

Association for Computational Linguistics 2026, Samyadeep Basu, Trung Bui, Hongjie Chen et autres

User simulation has long played a vital role in computer science due to its potential to support a wide range of applications. Language, as the primary medium of human communication, forms the foundation of social interaction and behavior. Consequently, simulating conversational behavior …

us (code pays fourni par la source)

Decomposition-Enhanced Training for Post-Hoc Attributions in Language Models

Association for Computational Linguistics 2026, Sriram Balasubramanian, Samyadeep Basu, Soheil Feizi et autres

Large language models (LLMs) are increasingly used for long-document question answering, where reliable attribution to sources is critical for trust. Existing post-hoc attribution methods work well for extractive QA but struggle in multi-hop, abstractive, and semi-extractive settings, where answers synthesize information across …

us (code pays fourni par la source)

From RAGs to rich parameters: Probing how language models utilize external knowledge over parametric information for factual queries

Hitesh Wadhwa, Rahul Seetharaman, Somyaa Aggarwal, Reshmi Ghosh et autres

Retrieval Augmented Generation (RAG) enriches the ability of language models to reason using external context to augment responses for a given user prompt. This approach has risen in popularity due to practical applications in various applications of language models in search, question/answering, …

us, gb (code pays fourni par la source)

Decomposition-Enhanced Training for Post-Hoc Attributions in Language Models

Sriram Balasubramanian, Samyadeep Basu, Koustava Goswami, Ryan A. Rossi et autres

Sriram Balasubramanian, Samyadeep Basu, Koustava Goswami, Ryan A. Rossi, Varun Manjunatha, Roshan Santhosh, Ruiyi Zhang, Soheil Feizi, Nedim Lipka. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

us (code pays fourni par la source)

A Survey on LLM-based Conversational User Simulation

Bo Ni, Yu Wang, Leyao Wang, Branislav Kveton et autres

Bo Ni, Yu Wang, Leyao Wang, Branislav Kveton, Franck Dernoncourt, Yu Xia, Hongjie Chen, Reuben Luera, Samyadeep Basu, Subhojyoti Mukherjee, Puneet Mathur, Nesreen K. Ahmed, Junda Wu, Li Li, Huixin Zhang, Ruiyi Zhang, Tong Yu, Sungchul Kim, Jiuxiang Gu, Zhengzhong Tu, Alexa …

SliderEdit: Continuous Image Editing with Fine-Grained Instruction Control

Arman Zarei, Samyadeep Basu, Mobina Pournemat, Sayan Nag et autres

Instruction-based image editing models have recently achieved impressive performance, enabling complex edits to an input image from a multi-instruction prompt. However, these models apply each instruction in the prompt with a fixed strength, limiting the user's ability to precisely and continuously control …

A Closer Look at Bias and Chain-of-Thought Faithfulness of Large (Vision) Language Models

Sriram Balasubramanian, Samyadeep Basu, Soheil Feizi

Chain-of-thought (CoT) reasoning enhances performance of large language models, but questions remain about whether these reasoning traces faithfully reflect the internal processes of the model. We present the first comprehensive study of CoT faithfulness in large vision-language models (LVLMs), investigating how both …

Mitigating Compositional Failures in Text-to-Image Models with Causal Text Embedding Refinement

Arman Zarei, Keivan Rezaei, Samyadeep Basu, Mehrdad Saberi et autres

Text-to-image diffusion-based generative models have the stunning ability to generate photo-realistic images and achieve state-of-the-art low FID scores on challenging image generation benchmarks. However, one of the primary failure modes of these text-to-image generative models is in composing attributes, objects, and their …

us (code pays fourni par la source)

A Survey on Mechanistic Interpretability for Multi-Modal Foundation Models

Zihao Lin, Samyadeep Basu, Mohammad Beigi, Varun Manjunatha et autres

The rise of foundation models has transformed machine learning research, prompting efforts to uncover their inner workings and develop more efficient and reliable applications for better control. While significant progress has been made in interpreting Large Language Models (LLMs), multimodal foundation models …