Related papers: VERIFY-RL: Verifiable Recursive Decomposition for …

As reinforcement learning (RL) achieves more success in solving complex tasks, more care is needed to ensure that RL research is reproducible and that algorithms herein can be compared easily and fairly with minimal bias. RL results are,…

Reinforcement Learning with Verifiable Rewards~(RLVR) has emerged as a powerful learn-to-reason paradigm for large reasoning models to tackle complex tasks. However, the current RLVR paradigm is still not efficient enough, as it works in a…

Understanding a Reinforcement Learning (RL) policy is crucial for ensuring that autonomous agents behave according to human expectations. This goal can be achieved using Explainable Reinforcement Learning (XRL) techniques. Although textual…

Formally verifying properties of software code has been a highly desirable task, especially with the emergence of LLM-generated code. In the same vein, they provide an interesting avenue for the exploration of formal verification and…

Language models are increasingly being used in important decision pipelines, so ensuring the correctness of their outputs is crucial. Recent work has proposed evaluating the "factuality" of claims decomposed from a language model generation…

Reinforcement learning (RL) is often credited with improving language model reasoning and generalization at the expense of degrading memorized knowledge. We challenge this narrative by observing that RL-enhanced models consistently…

Self-correction has emerged as a promising solution to boost the reasoning performance of large language models (LLMs), where LLMs refine their solutions using self-generated critiques that pinpoint the errors. This work explores whether…

Large language models (LLMs) have shown substantial capacity for generating fluent, contextually appropriate responses. However, they can produce hallucinated outputs, especially when a user query includes one or more false premises-claims…

We introduce Reasoning Core, a new scalable environment for Reinforcement Learning with Verifiable Rewards (RLVR), designed to advance foundational symbolic reasoning in Large Language Models (LLMs). Unlike existing benchmarks that focus on…

Several practical tools for automatically verifying functional programs (e.g., Liquid Haskell and Leon for Scala programs) rely on a heuristic based on unrolling recursive function definitions followed by quantifier-free reasoning using SMT…

Reasoning Large Language Models (R-LLMs) have significantly advanced complex reasoning tasks but often struggle with factuality, generating substantially more hallucinations than their non-reasoning counterparts on long-form factuality…

Reinforcement learning (RL) has recently emerged as a compelling approach for enhancing the reasoning capabilities of large language models (LLMs), where an LLM generator serves as a policy guided by a verifier (reward model). However,…

We present a new type system with support for proofs of programs in a call-by-value language with control operators. The proof mechanism relies on observational equivalence of (untyped) programs. It appears in two type constructors, which…

Reinforcement learning is critical to improving large reasoning models, but its success relies heavily on verifiable rewards (RLVR), making it hard to use in open-ended domains where correctness is ambiguous and cannot be verified.…

Reinforcement learning with verifiable rewards (RLVR) has demonstrated superior performance in enhancing the reasoning capability of large language models (LLMs). However, this accuracy-oriented learning paradigm often suffers from entropy…

Reinforcement Learning with Verifiable Rewards (RLVR) has become a prominent method for post-training Large Language Models (LLMs). However, verifiers are rarely error-free; even deterministic checks can be inaccurate, and the growing…

The emergence of compositional reasoning in large language models through reinforcement learning with verifiable rewards (RLVR) has been a key driver of recent empirical successes. Despite this progress, it remains unclear which…

Chain-of-Thought (CoT) prompting has become the de facto method to elicit reasoning capabilities from large language models (LLMs). However, to mitigate hallucinations in CoT that are notoriously difficult to detect, current methods such as…

The decompose-then-verify strategy for verification of Large Language Model (LLM) generations decomposes claims that are then independently verified. Decontextualization augments text (claims) to ensure it can be verified outside of the…

Hallucinations pose a challenge to the application of large language models (LLMs) thereby motivating the development of metrics to evaluate factual precision. We observe that popular metrics using the Decompose-Then-Verify framework, such…