Papers by Fuming Lai
MemBuilder: Reinforcing LLMs for Long-Term Memory Construction via Attributed Dense Rewards (2026.acl-long)
Copied to clipboard
| Challenge: | Memory-augmented frameworks fail to capture temporal evolution of historical states, limiting consistency in long-term dialogues. |
| Approach: | They propose a framework that trains models to orchestrate multi-dimensional memory construction with attributed dense rewards. |
| Outcome: | The proposed framework outperforms state-of-the-art closed-source models and generalizes well to OOD benchmarks. |
DART: Distilling Autoregressive Reasoning to Silent Thought (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing models that use Chain-of-Thought (CoT) have been slow to deploy in real-time applications due to its autoregressive nature. |
| Approach: | They propose a framework that replaces autoregressive CoT with non-autoregressive Silent Thought (ST) the framework uses a lightweight Reasoning Evolvement Module to align hidden states with the CoT pathway and a Reasoning Embedment Module (REM) during inference, only the ST pathway is activated, enabling the ST tokens to evolve into informative embeddings. |
| Outcome: | The proposed framework replaces autoregressive CoT with non-autoregressive Silent Thought (ST) it enables LLMs to generate answers directly from ST tokens without additional computational cost . |