Abductive Reasoning

Follow this topic Ask about this topic

Abductive reasoning works backward from observations to hypotheses that could explain them, supporting tasks such as diagnosis, causal inference, scientific discovery, and explaining model predictions. Approaches range from formal logic and knowledge-graph search to large language models that generate candidate explanations, often paired with symbolic verification, causal models, or reinforcement learning. Researchers evaluate these systems on tasks including event reasoning, commonsense benchmarks, knowledge-base completion, and minimal explanations for machine-learning decisions. A central challenge is producing explanations that are not only plausible but also valid, minimal, evidence-grounded, and robust to changes in wording or setting.

Latest papers

51 papers4 papers in the last four weeks, down 20% on the four weeks before. 0.0% of all new papers.

AbductionBench: Benchmarking Abductive Reasoning in Large Language Models
Nov 7, 2026Moein Salimi, Amir Qeysarbeigi, Babak Hosseini Mohtasham +8Abductive Reasoning Abductive Reasoning in Language Models
Sharif University of Technology · University of Tehran · Tarbiat Modares University
Learning to Bound: Monotonic, Calibrated Neural Upper Bounds for MPE Inference
Oct 29, 2026Akshay Vyas, Hailiang Dong, Shivvrat Arya +4Probabilistic Graphical Models Amortized Inference
The University of Texas at Dallas
Formal Explanations for Linear Temporal Logic over Finite Traces
Oct 26, 2026Valeria Fionda, Antonio Ielo, Francesco RiccaCounterfactual Explanations Linear Temporal Logic
University of Calabria, Italy
OntoAct: Hypothesis-Based Neuro-Symbolic Agents through Ontological Abstraction
Oct 15, 2026Donghyeok Lee, Bong Gyun Kang, Junyong Ahn +1Ontology-Based Reasoning Neuro-Symbolic Reasoning
Interdisciplinary Program in Artificial Intelligence, Seoul National University, Seoul, South Korea · Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea
ULTRADISCOVERY: Abductive Exploration in an Interconnected, Epistemically Open Universe
Oct 2, 2026Weihan Li, Tianshi Zheng, Yangqiu Song +2Autonomous Scientific Discovery Abductive Reasoning
Department of Mechanical Engineering, The University of Tokyo, Tokyo, Japan · Department of Computer Science and Engineering, HKUST, Hong Kong SAR, China · NVIDIA AI Technology Center (NVAITC), NVIDIA, Santa Clara, USA
Candidate Retention for Abductive Learning
Sep 30, 2026Hao-Yuan He, Yu Liu, Ming LiAbductive Reasoning Pseudo-Labeling
School of Artificial Intelligence, Nanjing University · National Key Laboratory for Novel Software Technology, Nanjing University
Abductive World Modeling via Causal Representation Learning
Sep 29, 2026Ziqi Liu, Songhan Yang, Linfan Zhou +4Causal Representation Learning Abductive Reasoning
BAIZ Team
Before the Arrest: Benchmarking LLMs on Criminal Profiling from Incomplete Evidence
Sep 17, 2026Yutong Yao, Yanjie Cao, Guanhua Chen +5Abductive Reasoning Legal Judgment Prediction
NLP2CT Lab, Faculty of Information Science and Computing, University of Macau · Faculty of Social Sciences, University of Macau
What Do Current Systematic Generalization Tasks Miss? A Reasoning-Centered Analysis
Sep 16, 2026Chengwen Qi, Deheng Ye, Yatao BianLogical Reasoning Inductive Inference
National University of Singapore · Nanyang Technological University
Solving Hard XAI Queries Based on a Compiled Dual-Rail Encoding
Sep 4, 2026Arthur Ledaguenel, Florent Capelli, Jean-Marie LagniezExplainable Artificial Intelligence Abductive Reasoning
CRIL, Université d’Artoi · LISIC, ULCO
DODR: Deterministic Operator-Driven Reasoning in Latent Space
Sep 4, 2026Weicai HuangLogical Reasoning Inductive Inference
Beijing MQPat Technologies, Co., Ltd.
Beyond $L_2$ : Generalizing Abductive Latent Explanations to Diverse Prototype-Based Architectures
Aug 17, 2026Jules Soria, Alban Grastien, Romain Xu-Darme +3Explainable Artificial Intelligence Neural Network Interpretability
Université Paris-Saclay, CEA, List, F-91120, Palaiseau, France
Evidence of Absence: Cross-Modal Abductive Risk Perception to Sustain World Models When Vision Fails
Aug 15, 2026Cong Xu, Ravi SankarAutonomous Driving Perception World Models for Autonomous Driving
iCONS Laboratory, Department of Electrical Engineering, University of South Florida, Tampa, FL 33620 USA
CEDAR-GRPO: Process-Aware Reinforcement Learning for General Abductive Reasoning in LLMs
Aug 14, 2026Moein Salimi, Danial Parnian, Shaygan Adim +6Abductive Reasoning in Language Models Abductive Reasoning
Sharif University of Technology · University of Tehran
LLMs Don't Pay for the Jump
Aug 14, 2026Paras Balani, Subhrakanta PandaAbductive Reasoning in Language Models Abductive Reasoning
Department of Mathematics and Department of Computer Science, Birla Institute of Technology and Science, Pilani, Hyderabad Campus, Jawahar Nagar, Kapra Mandal, Medchal District, Telangana 500078, India · Department of Computer Science, Birla Institute of Technology and Science, Pilani, Hyderabad Campus, Jawahar Nagar, Kapra Mandal, Medchal District, Telangana 500078, India
A Minimal $κ$ -- $τ$ Logic for Risk-Sensitive Abduction
Aug 8, 2026Remo PareschiAbductive Reasoning Causal Reasoning
STAKE Lab, University of Molise 86100 Campobasso, Italy
MoCA: Implicit Social Context Analysis
Aug 6, 2026Wenhao Xu, Kaiwen Zhang, Hao Li +10Abductive Reasoning Multimodal Large Language Models
National University of Singapore, Singapore, Singapore. · Wuhan University, Wuhan, China. · University of Oxford, Oxford, United Kingdom.
Adversarially Robust Abductive Fusion of Pre-trained Transformer-based Perception Models
Aug 4, 2026Mario Leiva, Yue Ma, Qinru Qiu +2Distribution Shift Abductive Reasoning
1DCIC, Universidad Nacional del Sur (UNS) & ICIC (UNS-CONICET), Bahía Blanca, Argentina · 2Syracuse University, Syracuse, NY USA
Abduction Without a Body? Representational Grounding and the Abduction Loop for Scientific Hypothesis Generation
Aug 3, 2026Michael FarmerAbductive Reasoning Scientific Hypothesis Generation
August 2026 (v1.18)
Alternating Abduction and Planning over Temporal Dynamic Knowledge Graphs
Jul 29, 2026Amir Hossein Sharafi, Alireza ShahbaziLogical Reasoning Abductive Reasoning
Najm · Department of Mathematics, Tafresh University
Representative Sets in Propositional Abduction
Jul 23, 2026Johannes Schmidt, Mohamed Maizia, Victor Lagerkvist +1Abductive Reasoning Nonmonotonic Reasoning
Jönköping University · Jönköping University, Linköping University · Linköping University
NeurOWL: An LLM-Based Neural-symbolic Framework for Incomplete OWL Ontology Reasoning
Jul 17, 2026Hui Yang, Jiaoyan Chen, Yiping Song +2Logical Reasoning Abductive Reasoning
The University of Manchester · Zhejiang University
Analytic Abduction: Causal Decomposition and Governed Commitment for Human--AI Coordination
Jul 16, 2026Remo PareschiAbductive Reasoning Causal Attribution
Stake Lab, University of Molise, Italy
Abductive Corroboration of Probabilistic AI Models for Forensic Synthetic Media Detection
Jul 3, 2026Junade AliAI-Assisted Decision Making Abductive Reasoning
The Alan Turing Institute Defence and National Security Programme London, United Kingdom
AGM-like Paraconsistent Partial Meet Abductive Expansion Operation
Jun 30, 2026Ulisses Franceschi ElianoBelief Revision Logical Reasoning
Institute of Philosophy and the Humanities (IFCH) and Centre for Logic, Epistemology and The History of Science (CLE), University of Campinas (UNICAMP)
Neurosymbolic Clinical Trial Matching via LLM-Driven Abduction and Logical Verification
Jun 18, 2026Baiyang Qu, Leonardo Ranaldi, Xi Wang +1Abductive Reasoning Neuro-Symbolic Reasoning
Department of Computer Science, University of Leicester, UK · School of Informatics, University of Edinburgh, UK · School of Computer Science, University of Sheffield, UK
The More the Merrier: Combining Properties for ABox Abduction under Repair Semantics in ELbot
Jun 17, 2026Anselm Haak, Patrick Koopmann, Yasir Mahmood +1Abductive Reasoning
Knowledge Representation Group, Paderborn University, Germany · Knowledge in Artificial Intelligence, Vrije Universiteit Amsterdam, The Netherlands · Data Science Group, Paderborn
DeFAb: A Dataset for Grounded Scientific Discovery through Defeasible Abduction
Jun 17, 2026Patrick Cooper, Avyay Sadhu, Alvaro VelasquezSymbolic Reasoning LLM Evaluation
University of Colorado Boulder
Beyond Explaining Predictions: Logic-Based Explanations for Confidence in Machine Learning Models
Jun 9, 2026Vinícius Peixoto Chagas, Carlos Henrique Leitão Cavalcante, Thiago Alves RochaExplainable Artificial Intelligence Abductive Reasoning
Abduction Prover in Isabelle/HOL
Jun 3, 2026Yutaka Nagashima, Daniel Sebastian GocAutomated Theorem Proving Abductive Reasoning
Institute of Computer Science, the Czech Academy of Sciences, Czechia

Nov 7, 2026J/K move · Enter open · S save

AbductionBench: Benchmarking Abductive Reasoning in Large Language Models

Moein Salimi, Amir Qeysarbeigi, Babak Hosseini Mohtasham, Faraz Doagoye Tehrani+7

Sharif University of Technology · University of Tehran · Tarbiat Modares University

Abductive Reasoning Abductive Reasoning in Language Models Language Model Reasoning Evaluation Adaptive Reasoning in Language Models

Open paper AskPDF arXiv

Abductive reasoning is the inference of plausible explanations for observed evidence. Evaluating this capability requires more than a single task: a system may need to generate an explanation, discriminate among competing hypotheses, revise its hypotheses as evidence arrives, or decide what evidence to seek next. Yet these capabilities are typically evaluated through isolated benchmarks with different task formulations and scoring procedures, so claims that a model is ``good at abduction'' may reflect only a narrow aspect of this capability. We introduce AbductionBench, a unified framework whose taxonomy characterizes abductive tasks. We instantiate this framework on 27 datasets spanning ten domains and evaluate eight LLMs from four model families evaluated on single-choice selection tasks. Beyond harmonizing heterogeneous tasks, AbductionBench separates final-answer evaluation from analysis of the observable reasoning traces and interaction histories produced alongside those answers. We analyze written traces as observable model outputs rather than faithful representations of internal computation, characterizing evidence use, hypothesis comparison, uncertainty, contradiction handling, and answer commitment. Jev matches or exceeds small LLMs when selecting among explanations over compact evidence but falls substantially behind on longer cases, suggesting that candidate discrimination and extended evidence integration place different demands on a system. In interactive tasks, the proportion of actions judged relevant declines at later steps, and a relevant first question is associated with a correct final answer; in the passive task, models rarely revise their initial hypothesis as new evidence arrives. Together, these findings highlight the value of evaluating abductive reasoning through both final answers and observable evidence use, hypothesis revision, and information gathering.