10 papers
Make Each Token Count: Towards Improving Long-Context Performance with KV Cache Eviction
Ngoc Bui, Hieu Trung Nguyen, Arman Cohan +1
The key-value (KV) cache is a major bottleneck in long-context inference, where memory and computation grow with sequence length. Existing KV eviction methods reduce this cost but…
Adaptive Rollout Allocation for Online Reinforcement Learning with Verifiable Rewards
Hieu Trung Nguyen, Bao Nguyen, Wenao Ma +3
Sampling efficiency is a key bottleneck in reinforcement learning with verifiable rewards. Existing group-based policy optimization methods, such as GRPO, allocate a fixed number o…
Retrospective Feature Estimation for Continual Learning
Nghia D. Nguyen, Hieu Trung Nguyen, Ang Li +3
The intrinsic capability to continuously learn a changing data stream is a desideratum of deep neural networks (DNNs). However, current DNNs suffer from catastrophic forgetting, wh…
Reasoning Planning for Language Models
Bao Nguyen, Hieu Trung Nguyen, Ruifeng She +2
Selecting an appropriate reasoning method for a given query remains a key challenge in language model generation. Existing approaches typically generate multiple candidate response…
Structured Pruning for Diverse Best-of-N Reasoning Optimization
Hieu Trung Nguyen, Bao Nguyen, Viet Anh Nguyen
Model pruning in transformer-based language models, traditionally viewed as a means of achieving computational savings, can enhance the model's reasoning capabilities. In this work…
Mixture-of-Personas Language Models for Population Simulation
Ngoc Bui, Hieu Trung Nguyen, Shantanu Kumar +4
Advances in Large Language Models (LLMs) paved the way for their emerging applications in various domains, such as human behavior simulations, where LLMs could augment human-genera…