Picture for Huayang Li

Huayang Li

ALR$^2$: A Retrieve-then-Reason Framework for Long-context Question Answering

Add code
Oct 04, 2024
Viaarxiv icon

On the Transformations across Reward Model, Parameter Update, and In-Context Prompt

Add code
Jun 24, 2024
Viaarxiv icon

M3T: A New Benchmark Dataset for Multi-Modal Document-Level Machine Translation

Add code
Jun 12, 2024
Viaarxiv icon

Cross-lingual Contextualized Phrase Retrieval

Add code
Mar 25, 2024
Viaarxiv icon

Inferflow: an Efficient and Highly Configurable Inference Engine for Large Language Models

Add code
Jan 16, 2024
Viaarxiv icon

GPT4Video: A Unified Multimodal Large Language Model for lnstruction-Followed Understanding and Safety-Aware Generation

Add code
Nov 25, 2023
Viaarxiv icon

Repetition In Repetition Out: Towards Understanding Neural Text Degeneration from the Data Perspective

Add code
Oct 16, 2023
Viaarxiv icon

TextBind: Multi-turn Interleaved Multimodal Instruction-following in the Wild

Add code
Sep 19, 2023
Viaarxiv icon

PandaGPT: One Model To Instruction-Follow Them All

Add code
May 25, 2023
Figure 1 for PandaGPT: One Model To Instruction-Follow Them All
Figure 2 for PandaGPT: One Model To Instruction-Follow Them All
Figure 3 for PandaGPT: One Model To Instruction-Follow Them All
Figure 4 for PandaGPT: One Model To Instruction-Follow Them All
Viaarxiv icon

A Frustratingly Simple Decoding Method for Neural Text Generation

Add code
May 22, 2023
Viaarxiv icon