Related papers: Composition Collapse: Stable Factual Knowledge Doe…

Despite remarkable advances, large language models often fail at compositional reasoning tasks, a phenomenon exemplified by the ``curse of two-hop reasoning''. This paper introduces the Identity Bridge, a simple yet powerful mechanism that…

Multi-hop question answering faces substantial challenges due to data sparsity, which increases the likelihood of language models learning spurious patterns. To address this issue, prior research has focused on diversifying question…

Chain-of-thought (CoT) traces are increasingly used both to improve language model capability and to audit model behavior, implicitly assuming that the visible trace remains synchronized with the computation that determines the answer. We…

RL training of multi-turn LLM agents is inherently unstable, and reasoning quality directly determines task performance. Entropy is widely used to track reasoning stability. However, entropy only measures diversity within the same input,…

An important open question in the use of large language models for knowledge-intensive tasks is how to effectively integrate knowledge from three sources: the model's parametric memory, external structured knowledge, and external…

Standard Transformers have a fixed computational depth, fundamentally limiting their ability to generalize to tasks requiring variable-depth reasoning, such as multi-hop graph traversal or nested logic. We propose a depth-recurrent…

This paper presents a novel framework for reconstructing multi-hop explanations in science Question Answering (QA). While existing approaches for multi-hop reasoning build explanations considering each question in isolation, we propose a…

Multi-hop reading comprehension (RC) questions are challenging because they require reading and reasoning over multiple paragraphs. We argue that it can be difficult to construct large multi-hop RC datasets. For example, even highly…

Despite the rapid progress in multihop question-answering (QA), models still have trouble explaining why an answer is correct, with limited explanation training data available to learn from. To address this, we introduce three explanation…

Fact-checking pipelines increasingly adopt the Decompose-Then-Verify paradigm, where texts are broken down into smaller claims for individual verification and subsequently combined for a veracity decision. While decomposition is…

Structured claim decomposition is often proposed as a solution for verifying complex, multi-faceted claims, yet empirical results have been inconsistent. We argue that these inconsistencies stem from two overlooked bottlenecks: evidence…

Due to the lack of insufficient data, existing multi-hop open domain question answering systems require to effectively find out relevant supporting facts according to each question. To alleviate the challenges of semantic factual sentences…

Automatic multi-hop fact verification task has gained significant attention in recent years. Despite impressive results, these well-designed models perform poorly on out-of-domain data. One possible solution is to augment the training data…

The emergence of reasoning models and their integration into practical AI chat bots has led to breakthroughs in solving advanced math, deep search, and extractive question answering problems that requires a complex and multi-step thought…

The fine-tuning of deep pre-trained models has revealed compositional properties, with multiple specialized modules that can be arbitrarily composed into a single, multi-task model. However, identifying the conditions that promote…

Symbolic logical reasoning is a critical yet underexplored capability of large language models (LLMs), providing reliable and verifiable decision-making in high-stakes domains such as mathematical reasoning and legal judgment. In this…

Human cognition exhibits systematic compositionality, the algebraic ability to generate infinite novel combinations from finite learned components, which is the key to understanding and reasoning about complex logic. In this work, we…

We present ACCORD, a framework and benchmark suite for disentangling the commonsense grounding and reasoning abilities of large language models (LLMs) through controlled, multi-hop counterfactuals. ACCORD introduces formal elements to…

While Chain-of-Thought (CoT) prompting has become a cornerstone for complex reasoning in Large Language Models (LLMs), the faithfulness of the generated reasoning remains an open question. We investigate the Decoupling Hypothesis: that…

Post-trained language models produce less varied outputs than their base counterparts. This output diversity collapse undermines inference-time scaling methods that rely on varied samples, and risks homogenizing model outputs on creative…