Logical Reasoning of Large Language Models

The Fourteenth International Conference on Learning Representations

April 26th, 2026

Riocentro Convention and Event Center, Rio de Janeiro, Brazil

Instruction

Large language models (LLMs) have achieved remarkable breakthroughs in natural language understanding and generation, but their logical reasoning capabilities remain a significant bottleneck. Logical reasoning is crucial for tasks requiring precise deduction, induction, or abduction, such as medical diagnosis, legal reasoning, and scientific hypothesis verification. However, LLMs often fail to handle complex logical problems with multiple premises and constraints, and they frequently produce self-contradictory responses across different questions. These limitations not only restrict the reliability of LLMs in complex problem-solving but also hinder their real-world applications.

In response to these emerging needs, we propose the workshop on Logical Reasoning of LLMs. This workshop will explore the challenges and opportunities for improving deduction, induction, and abduction capabilities of LLMs, implementing symbolic representation and reasoning via LLMs, avoiding logical contradictions across responses to multiple related questions, enhancing LLM reasoning by leveraging external logical solvers, and benchmarking LLM logical reasoning and consistencies. As LLMs continue to expand their role in AI research and applications, this workshop will serve as a platform to discuss and refine the methods for advancing logical reasoning within LLMs.

Topics

Topics of interest include, but are not limited to:

• • Deduction, Induction, and Abduction Capabilities of LLMs;

  • Symbolic Representation and Reasoning;

  • Avoiding Logical Contradictions Across Responses to Multiple Related Questions;

  • Enhancing LLM Reasoning by Leveraging External Logical Solvers;

  • Benchmarks and Evaluation for Logical Reasoning and Consistencies.

Important Dates

• First-Round Submission Deadline: February 20, 2026

• First-Round Notification of Acceptance: March 1, 2026 

• Second-Round Submission Deadline: March 20, 2026

• Second-Round Notification of Acceptance: March 31, 2026 

All deadlines are specified in Anywhere on Earth (AoE). The Second-Round submissions are Not Available to ICLR 2026 Financial Assistance and Volunteering.

Submission Guidelines

We welcome two types of papers:

• Regular papers: Full-length research papers from 4 to 10 pages (excluding references and supplementary material);

• Tiny/Short papers: research/position papers from 2 to 4 pages (excluding references and supplementary material).

Papers should be submitted in the ICLR 2026 format (https://github.com/ICLR/Master-Template/raw/master/iclr2026.zip). 

The review process will be double-blinded; therefore, all identifying information, including author names, affiliations, and acknowledgments, must be excluded from submissions. We welcome both unpublished manuscripts and previously accepted or published papers, provided that their submission does not violate any dual-submission policies. The contributions are non-archival and all accepted papers will be made publicly available on OpenReview. 

There will be Oral Presentation Awards, Outstanding Paper Awards, and one Best Paper Award for accepted papers with outstanding quality.

Financial Assistance for Author Attendance

Our workshop is open to researchers, practitioners, and industry professionals interested in logical reasoning of large language models. There are no specific criteria or maximum number of attendees.

Authors of accepted papers in our workshop, including tiny/short papers, may be eligible for funding from ICLR to attend in person or virtually, but need to submit a separate application for Financial Assistance that is used by the DEI Chairs to make funding determinations for all ICLR contributors. This application for Finalcial Assistance will close on March 2nd, and more details can be found in https://iclr.cc/Conferences/2026/FinancialAssistance.

Reviewer Recruitment

We are actively seeking reviewers to join our program committee. If you are interested in contributing to the peer-review process, please contact Fengxiang Cheng at f.cheng@uva.nl.

Program Chairs

Student Assistants

Program Committee Members

Outstanding Reviewers:

Shabbir Hussain (Amazon), Hari Charan Panjwani (Facebook), Jerry Wang (National Chengchi University), Kartik Balasubramaniam (CapitalOne), Aditya Gaydhani (Target Corporation), Suguna Varshini Velury (NVIDIA), Miraj Shah (University of Pennsylvania), Anushree Sinha (Google), Shaashwat Agrawal (Northwestern University), Marc Boubnovski Martell (Novo Nordisk), Mandakinee Singh Patel (Apple).

Reviewers:

Tianye Song (Synopsys Inc.), Sheshananda Reddy Kandula (IEEE), Aprameya Bharadwaj (Adobe Systems), Arman Engin Sucu (Northeastern University), Raphi Kang (California Institute of Technology), Dan Le (Google Deepmind), Varun Kotte (Adobe Systems), Ajay Krishna Borra (SalesForce), Hongqiao Chen (California Institute of Technology), Karun Thankachan (Walmart), Bob Wei (Microsoft), Leander Girrbach (Technische Universität München), Taigao Ma (University of Michigan - Ann Arbor), Md Muntaqim Meherab (Daffodil International University), Hehuan Liu (IEG), Srirangan Madhavan (University of California, San Diego), Olivia Peiyu Wang (University of California, Santa Cruz), Tai Vu (OpenAI), Kunal Mulwani (Google), Yashu Pura Lingaraju (Stripe Inc.), Pratik Sridatt Jayarao (Amazon), Vethavikashini Chithrra Raghuram (Amazon), Sri krishna rao achyutuni (ZS Associates), Thejas Venkatesh (Samaya AI, Inc.), Xianglong Hu (Amazon), Tao Jiang (University of the Chinese Academy of Sciences), Nischal Reddy Chandra (Adobe Systems), Dhivya Nagasubramanian (University of Minnesota - Twin Cities), Md Sultanul Islam Ovi (George Mason University), Rachana Ramchandar (Pes University), Mayank Ravishankara, Savan Doshi (Arizona State University), Chuchu Wu (Deloitte Consulting), Raghav Sharma (Workday), Alok Upadhyay (Amazon), Rohith Uppala (LinkedIn), Vasanth Kalingeri (Facebook), Harshitha Ravindra (Amazon), Vimanyu Taneja (University of Cambridge), Anshu Rajendra (Uber), Sanket Badhe (Google), Dhyey Dharmendrakumar Mavani (LinkedIn & Amherst College), Ashwin Aravind (AgentWall), SHIH-CHIH LIN (National Tsinghua University).

Sponsorship Packages

Diamond ($10,000) :

• (5) complimentary registrations (the Workshop registration)

• One keynote talk slot from the sponsored company

• Obtain the contact information of the Workshop attendees upon their approval

• Exclusive sponsorship of one Conference Award; 2 options include:

(a) Best Paper Award

(b) Outstanding Paper Award

• Host an invitation only evening reception inclusive of private reception space and AV/Technical support; sponsor responsible for food and beverage

• Acknowledgement as a ICLR Corporate Affiliate on the website and in the published proceeding

• Opportunity to exhibit complimentary front booth space (participation optional)

• Discounted opportunities for add-on elements to increase visibility and reach

Platinum ($5,000) :

• (3) complimentary registrations (the Workshop registration)

• Exclusive sponsorship of one Conference Award; 2 options include:

(a) Best Paper Award

(b) Outstanding Paper Award

• Host an invitation only evening reception inclusive of private reception space and AV/Technical support; sponsor responsible for food and beverage

• Acknowledgement as a ICLR Corporate Affiliate on the ICLR Workshop website and in the published proceeding

• Opportunity to exhibit complimentary front booth space (participation optional)

• Discounted opportunities for add-on elements to increase visibility and reach

Please contact us at hxli@stu.pku.edu.cn for additional available Diamond and Platinum Selections.

Oral Presentation

Learning Reasoning Reward Models from Expert Demonstration via Inverse Reinforcement Learning

Claudio Fanconi, Nicolás Astorga, Mihaela van der Schaar

RAG over Tables: Hierarchical Memory Index, Multi-Stage Retrieval, and Benchmarking

Jiaru Zou, Dongqi Fu, Sirui Chen, Xinrui He, Zihao Li, Yada Zhu, Jiawei Han, Jingrui He

Position: Logical Reasoning Evaluation Must Measure Social Bias

Sofía Martinelli, Guido Ivetta, Luciana Benotti

Beyond Clause Count: A Study of Proof-Relevant Difficulty in LLM SAT Reasoning

Tao Jiang, Shaowei Cai

Actor-Curator: Co-adaptive curricula via policy-improvement bandits for post-training

Zhengyao Gu, Jonathan Light, Raul Astudillo, Ziyu Ye, Langzhou He, Wei Cheng, Santiago Paternain, Philip S. Yu, Yisong Yue

Revisiting Causal Reasoning in Language Models through Controlled Synthetic Worlds

Abhirath Sangala, Amit Sharma, Vineeth N. Balasubramanian

From Natural Language to Exact Cover: A Neuro-Symbolic Approach to Zebra Puzzles

Paulius Skaisgiris, Thomas Pammer, Veronika Semmelrock, Mykyta Ielanskyi, Maximilian Heisinger, Erich Kobler

Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty

Baishali Chaudhury, Mengdie Flora Wang, Hyunji Hayley Park, Rahul Ghosh, Sungmin Hong, Jae Oh Woo

Program Schedule

9:00am – 9:05am Welcome Remarks

9:05am – 10:00am Tutorial

10:00am – 10:30am Break

10:30am – 10:50am Oral 1: Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty

10:50pm – 11:30pm Keynote 1

11:30am – 11:50am Poster Session 1

11:50am – 12:10pm Oral 2: Actor-Curator: Co-adaptive curricula via policy-improvement bandits for post-training 

12:10pm – 12:30pm Oral 3: Learning Reasoning Reward Models from Expert Demonstration via Inverse Reinforcement Learning 

12:30pm – 1:45pm Lunch 

1:45pm – 2:05pm Oral 4: RAG over Tables: Hierarchical Memory Index, Multi-Stage Retrieval, and Benchmarking

2:05pm – 2:45pm Keynote 2  

2:45pm – 3:15pm Break 

3:15pm – 3:35pm Oral 5: From Natural Language to Exact Cover: A Neuro-Symbolic Approach to Zebra Puzzles 

3:35pm – 3:55pm Oral 6: Revisiting Causal Reasoning in Language Models through Controlled Synthetic Worlds 

3:55pm – 4:15pm Poster Session 2 

4:15pm – 4:35pm Oral 7: Beyond Clause Count: A Study of Proof-Relevant Difficulty in LLM SAT Reasoning 

4:35pm – 4:55pm Oral 8: Logical Reasoning Evaluation and Social Bias

4:55pm – 5:00pm Closing Remarks