16 citations · 36 across the 34 of their papers we have counts for
36 papers
DISCO: Distributed Long Context Scaling with Grounding-Reasoning Disaggregation
Guanzheng Chen, Viet Dac Lai, Subhojyoti Mukherjee +5
While Large Language Models (LLMs) advertise million-token context windows, reasoning quality often collapses as inputs grow -- a phenomenon termed context rot. This failure stems…
MoMHa: Multi-Objective Optimization of LLM Harnesses over Accuracy, Safety, and Tokens
Subhojyoti Mukherjee, Md Mehrab Tanjim
Most work on improving large language models treats accuracy as the sole objective. We argue that the harness, the Python code surrounding the model that constructs prompts, routes…
KnowSim: Evaluating Information Calibration in LLM Assistants with User Simulators that Learn
Yoonjoo Lee, Hyoungwook Jin, Tae Soo Kim +3
To effectively collaborate with users on knowledge-intensive tasks, Large Language Models (LLMs) must perform information calibration: matching content to a user's evolving underst…
AdvantageFlow: Regularized Advantage-Weighted RL in Flow Models
Branislav Kveton, Anup Rao, Subhojyoti Mukherjee +2
We present AdvantageFlow, a forward-process reinforcement learning (RL) algorithm for rectified flow models. The algorithm minimizes an advantage-weighted prediction loss, which ma…
MOCHA: Multi-Objective Chebyshev Annealing for Agent Skill Optimization
Md Mehrab Tanjim, Jayakumar Subramanian, Xiang Chen +6
LLM agents organize behavior through skills -- structured natural-language specifications that govern how an agent reasons, retrieves, and responds. Unlike monolithic prompts, skil…
Sparse Personalized Text Generation with Multi-Trajectory Reasoning
Bo Ni, Haowei Fu, Qinwen Ge +10
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 d…
Stepwise Credit Assignment for GRPO on Flow-Matching Models
Yash Savani, Branislav Kveton, Yuchen Liu +5
Flow-GRPO successfully applies reinforcement learning to flow models, but uses uniform credit assignment across all steps. This ignores the temporal structure of diffusion generati…
Agentic Planning with Reasoning for Image Styling via Offline RL
Subhojyoti Mukherjee, Stefano Petrangeli, Branislav Kveton +3
Direct prompt-based editing often fails on complex transformations because vague and subjective prompts often require nuanced understanding of what should be changed in the image.…
Partial Policy Gradients for RL in LLMs
Puneet Mathur, Branislav Kveton, Subhojyoti Mukherjee +1
Reinforcement learning is a framework for learning to act sequentially in an unknown environment. We propose a natural approach for modeling policy structure in policy gradients. T…
InfinityStory: Unlimited Video Generation with World Consistency and Character-Aware Shot Transitions
Mohamed Elmoghany, Liangbing Zhao, Xiaoqian Shen +27
Generating long-form storytelling videos with consistent visual narratives remains a significant challenge in video synthesis. We present a novel framework, dataset, and a model th…
Human-Aligned MLLM Judges for Fine-Grained Image Editing Evaluation: A Benchmark, Framework, and Analysis
Runzhou Liu, Hailey Weingord, Sejal Mittal +18
Evaluating image editing models remains challenging due to the coarse granularity and limited interpretability of traditional metrics, which often fail to capture aspects important…
Reasoning-Based Personalized Generation for Users with Sparse Data
Bo Ni, Branislav Kveton, Samyadeep Basu +14
Large Language Model (LLM) personalization holds great promise for tailoring responses by leveraging personal context and history. However, real-world users usually possess sparse…
Learning to Reason in LLMs by Expectation Maximization
Junghyun Lee, Branislav Kveton, Anup Rao +4
Large language models (LLMs) solve reasoning problems by first generating a rationale and then answering. We formalize reasoning as a latent variable model and derive a reward-base…
StreamGaze: Gaze-Guided Temporal Reasoning and Proactive Understanding in Streaming Videos
Daeun Lee, Subhojyoti Mukherjee, Branislav Kveton +6
Streaming video understanding requires models not only to process temporally incoming frames, but also to anticipate user intention for realistic applications such as Augmented Rea…
MLLM as a UI Judge: Benchmarking Multimodal LLMs for Predicting Human Perception of User Interfaces
Reuben A. Luera, Ryan Rossi, Franck Dernoncourt +12
In an ideal design pipeline, user interface (UI) design is intertwined with user research to validate decisions, yet studies are often resource-constrained during early exploration…
A Survey on Long-Video Storytelling Generation: Architectures, Consistency, and Cinematic Quality
Mohamed Elmoghany, Ryan Rossi, Seunghyun Yoon +26
Despite the significant progress that has been made in video generative models, existing state-of-the-art methods can only produce videos lasting 5-16 seconds, often labeled "long-…
Offline RL by Reward-Weighted Fine-Tuning for Conversation Optimization
Subhojyoti Mukherjee, Viet Dac Lai, Raghavendra Addanki +6
Offline reinforcement learning (RL) is a variant of RL where the policy is learned from a previously collected dataset of trajectories and rewards. In our work, we propose a practi…
A Personalized Conversational Benchmark: Towards Simulating Personalized Conversations
Li Li, Peilin Cai, Ryan A. Rossi +21
We present PersonaConvBench, a large-scale benchmark for evaluating personalized reasoning and generation in multi-turn conversations with large language models (LLMs). Unlike exis…
From Selection to Generation: A Survey of LLM-based Active Learning
Yu Xia, Subhojyoti Mukherjee, Zhouhang Xie +31
Active Learning (AL) has been a powerful paradigm for improving model efficiency and performance by selecting the most informative data points for labeling and training. In recent…
Logits are All We Need to Adapt Closed Models
Gaurush Hiranandani, Haolun Wu, Subhojyoti Mukherjee +1
Many commercial Large Language Models (LLMs) are often closed-source, limiting developers to prompt tuning for aligning content generation with specific applications. While these m…
Multi-Objective Alignment of Large Language Models Through Hypervolume Maximization
Subhojyoti Mukherjee, Anusha Lalitha, Sailik Sengupta +2
Multi-objective alignment from human feedback (MOAHF) in large language models (LLMs) is a challenging problem as human preferences are complex, multifaceted, and often conflicting…
Off-Policy Evaluation from Logged Human Feedback
Aniruddha Bhargava, Lalit Jain, Branislav Kveton +2
Learning from human feedback has been central to recent advances in artificial intelligence and machine learning. Since the collection of human feedback is costly, a natural questi…
SaVeR: Optimal Data Collection Strategy for Safe Policy Evaluation in Tabular MDP
Subhojyoti Mukherjee, Josiah P. Hanna, Robert Nowak
In this paper, we study safe data collection for the purpose of policy evaluation in tabular Markov decision processes (MDPs). In policy evaluation, we are given a \textit{target}…
Pretraining Decision Transformers with Reward Prediction for In-Context Multi-task Structured Bandit Learning
Subhojyoti Mukherjee, Josiah P. Hanna, Qiaomin Xie +1
We study learning to learn for the multi-task structured bandit problem where the goal is to learn a near-optimal algorithm that minimizes cumulative regret. The tasks share a comm…