Large Language Model
WISE: Rethinking the Knowledge Memory for Lifelong Model Editing of Large Language Models
Where the updated knowledge resides in memories is a fundamental question for model editing.
Accelerating Greedy Coordinate Gradient and General Prompt Optimization via Probe Sampling
Touvron et al., 2023; Jiang et al., 2023) has become a central theme of research.
Unelicitable Backdoors in Language Models via Cryptographic Transformer Circuits
In this paper, we introduce a novel class of backdoors in transformer models, that, in contrast to prior art, are unelicitable in nature.
Autoformalizing Mathematical Statements by Symbolic Equivalence and Semantic Consistency Zenan Li1 Yifan Wu2 Zhaoyu Li3 Xinming Wei 2
This work was partially done during Zenan's and Yifan's internships at MSRA. 38th Conference on Neural Information Processing Systems (NeurIPS 2024).
MACM: Utilizing a Multi-Agent System for Condition Mining in Solving Complex Mathematical Problems Bin Lei
Recent advancements in large language models, such as GPT -4, have demonstrated remarkable capabilities in processing standard queries.
LRM-Zero: Training Large Reconstruction Models with Synthesized Data
Unlike previous 3D datasets (e.g., Objaverse) which are often captured or crafted by humans to approximate real 3D data, Zeroverse completely ignores realistic global semantics but is
Understanding Emergent Abilities of Language Models from the Loss Perspective
In this paper, we propose to study emergent abilities in the lens of pre-training loss, instead of model size or training compute.
