collaborators

5 papers

cs.CL2026

SeDeM: Selective Decompression of Hidden-State Memories for Long-Context Question Answering

Maryam Haghifam, Jason Cong, Yizhou Sun

Long-context inference with large language models (LLMs) is costly: self-attention during prefill scales quadratically with sequence length, and the key-value (KV) cache grows with…

cs.CL2026

HMT: Semantic Hierarchy-Aware Hierarchical Memory Transformer

Maryam Haghifam, Zifan He, Jason Cong +1

Transformer-based LLMs achieve strong results on many language tasks; however, long inputs remain challenging because context windows are finite, and prefill latency and memory gro…

cs.CL2026

Structure-Aware RAG: Structured Retrieval Augmented Generation from Noisy Data for Conversational Agents

Kaiqiao Han, LuAn Tang, Renliang Sun +6

Large Language Models (LLMs) have been widely adopted in conversational applications. However, their reliance on parametric knowledge limits reliability in real-world scenarios tha…

cs.CL2026

Compressing Sequences in the Latent Embedding Space: -Token Merging for Large Language Models

Zihao Xu, John Harvill, Ziwei Fan +3

Large Language Models (LLMs) incur significant computational and memory costs when processing long prompts, as full self-attention scales quadratically with input length. Token com…

cs.CL2026

ConFu: Contemplate the Future for Better Speculative Sampling

Zongyue Qin, Raghavv Goel, Mukul Gagrani +3

Speculative decoding has emerged as a powerful approach to accelerate large language model (LLM) inference by employing lightweight draft models to propose candidate tokens that ar…