XLLM-Reason-Plan @ COLM 2025

11 papers

Top Research Directions

The 17 research directions with XLLM-Reason-Plan @ COLM 2025 papers, ranked by paper count. Share is the fraction of the edition’s papers in the direction.

#DirectionPapersShare
1Activation Patching for Mechanistic Interpretability218.2%
2Activation Steering for Inference Time Model Control19.1%
3Dynamic Inference Compute Allocation19.1%
4How Fine Tuning Reshapes Internal Model Circuitry19.1%
5How Language Models Collapse Linguistic Ambiguity19.1%
6Intermediate Reasoning Chains for Deliberative Inference19.1%
7Linear Concept Directions in Latent Spaces19.1%
8Mitigating the Inference Latency of Chain of Thought19.1%
9Overcoming the Limits of Multi Step Language Model Reasoning19.1%
10Projecting Intermediate Transformer States to the Vocabulary Space19.1%
11Recurrent Weight Sharing for Parameter Efficient Depth19.1%
12Scaling In Context Task Representations to Complex Reasoning19.1%
13Schema Linking for Text to SQL Parsing19.1%
14Systematic Biases in Large Language Model Evaluators19.1%
15The Cross-Genre Generalization Gap19.1%
16The Language Bias of English Centric Benchmarks19.1%
17Understanding Depth Saturation in Deep Neural Networks19.1%

Download all directions as CSV

Institutions

Institutions credited by XLLM-Reason-Plan @ COLM 2025 papers: a paper credits every institution named in its affiliations, once. 17 institutions share 8 credited papers.

#InstitutionPapersShare
1Harvard University 225.0%
2McGill University 225.0%
3Mila - Quebec Artificial Intelligence Institute 225.0%
4Stanford University 225.0%
5University of Washington 225.0%
603dgzv892 112.5%
7Brown University 112.5%
8Canadian Institute for Advanced Research 112.5%
9Hong Kong University of Science and Technology 112.5%
10Huawei Technologies (China) 112.5%
11Royal Bank of Canada 112.5%
12The Hong Kong University of Science and Technology (Guangzhou) 112.5%
13Tsinghua University 112.5%
14University of Alberta 112.5%
15University of California, Los Angeles 112.5%
16University of Edinburgh 112.5%
17University of Illinois Urbana-Champaign 112.5%

Download all institutions as CSV

About this data

Papers. XLLM-Reason-Plan @ COLM 2025 lists 11 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 8 of 11 papers (72.7%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-18 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “XLLM-Reason-Plan @ COLM 2025”, https://lacuna.tiptreesystems.com/venue/xllm-reason-plan-colm/2025.

Top Authors

AuthorPapers
Ranjay Krishna2
Abbas Ghaddar1
Bin Wang1
Damien Graux1
Enqi Liu1
Ge Zhang1
Guangyi Liu1
Helena Vasconcelos1
Himabindu Lakkaraju1
Hongzhe Du1
Idan Shenfeld1
Irina Saparina1
Isha Puri1
Jackie Chi Kit Cheung1
Jacob Andreas1
Jeff Z. Pan1
Jiaming Zhou1
Jianye HAO1
Karim Saraipour1
Kaustubh Vyas1

Impactful Papers

Papers from XLLM-Reason-Plan @ COLM 2025 that shaped research clusters.

Influential

Papers

Title
Are General-Purpose LLMs Ready for Planning? A Large- Scale Evaluation in PDDL
Before You 〈think/〉, Monitor: Implementing Flavell's Metacognitive Framework in LLMs
Can LLMs Reason Abstractly Over Math Word Problems Without CoT? Disentangling Abstract Formulation From Arithmetic Computation
Case-Based Reasoning Enhances the Predictive Power of LLMs in Drug-Drug Interaction
Disambiguate First, Parse Later: Generating Interpretations for Ambiguity Resolution in Semantic Parsing
Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Data
How Post-Training Reshapes LLMs: A Mechanistic View on Knowledge, Truthfulness, Refusal, and Confidence
Latent Chain-of-Thought? Decoding the Depth-Recurrent Transformer
ReCalibrate: RL for Uncertainty-Aware Reasoning in LLMs
Rethinking (Human) Preference Evaluation of LLM Rationales
Rethinking Human Preference Evaluation of LLM Rationales