Weibin Liao

activity

20242026

most cited3DS: Medical Domain Adaptation of LLMs via Decomposed Difficulty-based Data Selection

1 citations · 2 across the 9 of their papers we have counts for

collaborators

Showing cs.CLShow all

3 papers · 1 filter

cs.CL2026

The Imitation Game: When LLMs Learn to Reason Like Programs via Code-Centric Reasoning Data Synthesis

Jinyang Zhang, Weibin Liao, Keqin Bao +10

Large Language Models (LLMs) excel at programming tasks but frequently fail at deterministic, fine-grained reasoning in natural language, relying heavily on semantic approximations…

cs.CL2026

The Tell-Tale Norm: $ℓ_{2}$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models

Jinyang Zhang, Hongxin Ding, Yue Fang +4

Recent work has sought to understand Large Language Models (LLMs) reasoning, yet a principled, model-intrinsic signal that captures its layer-wise reasoning dynamics remains undere…

cs.CL2025★ 1 cited

LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models

Weibin Liao, Xin Gao, Tianyu Jia +6

Natural Language to SQL (NL2SQL) aims to translate natural language queries into executable SQL statements, offering non-expert users intuitive access to databases. While recent ap…

The Imitation Game: When LLMs Learn to Reason Like Programs via Code-Centric Reasoning Data Synthesis

The Tell-Tale Norm: ℓ2​ Magnitude as a Signal for Reasoning Dynamics in Large Language Models

LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models

The Tell-Tale Norm: $ℓ_{2}$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models