English
Related papers

Related papers: RORA: Robust Free-Text Rationale Evaluation

200 papers

The inherent capabilities of a language model (LM) and the reasoning strategies it employs jointly determine its performance in reasoning tasks. While test-time scaling is regarded as an effective approach to tackling complex reasoning…

Computation and Language · Computer Science 2025-05-27 Zhihong Pan , Kai Zhang , Yuze Zhao , Yupeng Han

Reinforcement learning has become a central paradigm for improving LLM reasoning, but most existing methods optimize policies over discrete token sequences. This creates a mismatch between the optimization space and the structure of…

Machine Learning · Computer Science 2026-05-19 Haoqiang Kang , Yizhe Zhang , Nikki Lijing Kuang , Yi-An Ma , Lianhui Qin

Formal explainability guarantees the rigor of computed explanations, and so it is paramount in domains where rigor is critical, including those deemed high-risk. Unfortunately, since its inception formal explainability has been hampered by…

Artificial Intelligence · Computer Science 2024-12-04 Xuanxiang Huang , Joao Marques-Silva

While test-time scaling has enabled large language models to solve highly difficult tasks, state-of-the-art results come at exorbitant compute costs. These inefficiencies can be attributed to the miscalibration of post-trained language…

Machine Learning · Computer Science 2026-04-02 Cai Zhou , Zekai Wang , Menghua Wu , Qianyu Julie Zhu , Flora C. Shi , Chenyu Wang , Ashia Wilson , Tommi Jaakkola , Stephen Bates

Reward learning enables the application of reinforcement learning (RL) to tasks where reward is defined by human judgment, building a model of reward by asking humans questions. Most work on reward learning has used simulated environments,…

Computation and Language · Computer Science 2020-01-10 Daniel M. Ziegler , Nisan Stiennon , Jeffrey Wu , Tom B. Brown , Alec Radford , Dario Amodei , Paul Christiano , Geoffrey Irving

Large Language Models (LLMs) rely on generating extensive intermediate reasoning units (e.g., tokens, sentences) to enhance final answer quality across a wide range of complex tasks. While this approach has proven effective, it inevitably…

Computation and Language · Computer Science 2025-06-05 Joonwon Jang , Jaehee Kim , Wonbin Kweon , Seonghyeon Lee , Hwanjo Yu

Large Reasoning Models (LRMs) have demonstrated a latent capacity for complex reasoning by spontaneously exhibiting cognitive behaviors such as step-by-step reasoning, reflection, and backtracking, commonly referred to as "Aha Moments".…

Artificial Intelligence · Computer Science 2025-08-07 Rui Ha , Chaozhuo Li , Rui Pu , Sen Su

We propose a framework for robust evaluation of reasoning capabilities of language models, using functional variants of benchmarks. Models that solve a reasoning test should exhibit no difference in performance over the static version of a…

Artificial Intelligence · Computer Science 2024-03-01 Saurabh Srivastava , Annarose M B , Anto P , Shashank Menon , Ajay Sukumar , Adwaith Samod T , Alan Philipose , Stevin Prince , Sooraj Thomas

The reasoning steps generated by LLMs might be incomplete, as they mimic logical leaps common in everyday communication found in their pre-training data: underlying rationales are frequently left implicit (unstated). To address this…

Artificial Intelligence · Computer Science 2025-06-17 Dongwei Jiang , Guoxuan Wang , Yining Lu , Andrew Wang , Jingyu Zhang , Chuyu Liu , Benjamin Van Durme , Daniel Khashabi

The capability to reason from text is crucial for real-world NLP applications. Real-world scenarios often involve incomplete or evolving data. In response, individuals update their beliefs and understandings accordingly. However, most…

Computation and Language · Computer Science 2024-10-18 Bryan Wilie , Samuel Cahyawijaya , Etsuko Ishii , Junxian He , Pascale Fung

Large language models (LMs) are capable of generating free-text rationales to aid question answering. However, prior work 1) suggests that useful self-rationalization is emergent only at significant scales (e.g., 175B parameter GPT-3); and…

Computation and Language · Computer Science 2024-05-24 Sahana Ramnath , Brihi Joshi , Skyler Hallinan , Ximing Lu , Liunian Harold Li , Aaron Chan , Jack Hessel , Yejin Choi , Xiang Ren

RL with Verifiable Rewards (RLVR) has emerged as a promising paradigm for improving the reasoning abilities of large language models (LLMs). Current methods rely primarily on policy optimization frameworks like PPO and GRPO, which follow…

Machine Learning · Computer Science 2025-09-30 Haoran He , Yuxiao Ye , Qingpeng Cai , Chen Hu , Binxing Jiao , Daxin Jiang , Ling Pan

Large language models can answer causal questions correctly for the wrong reasons. Current RL methods reward \emph{what} a model concludes but ignore \emph{why}, reinforcing correlational shortcuts -- a failure we call \emph{Reward…

Artificial Intelligence · Computer Science 2026-05-21 Edward Y. Chang , Longling Geng

As Large Reasoning Models (LRMs) are increasingly deployed, auditing their chain-of-thought (CoT) traces for safety becomes critical. Recent work has reported that monitorability--the degree to which CoT faithfully and informatively…

Artificial Intelligence · Computer Science 2026-02-05 Zidi Xiong , Shan Chen , Himabindu Lakkaraju

We introduce CLEAR (Contrasting Textual Feedback with Experts and Amateurs for Reasoning), a novel approach to language model reasoning that leverages the strengths of a larger (expert) model and smaller (amateur) model. The expert and…

Computation and Language · Computer Science 2025-04-11 Andrew Rufail , Daniel Kim , Sean O'Brien , Kevin Zhu

Despite the increasing effectiveness of language models, their reasoning capabilities remain underdeveloped. In particular, causal reasoning through counterfactual question answering is lacking. This work aims to bridge this gap. We first…

Computation and Language · Computer Science 2025-03-18 Alihan Hüyük , Xinnuo Xu , Jacqueline Maasch , Aditya V. Nori , Javier González

A new generation of AI models generates step-by-step reasoning text before producing an answer. This text appears to offer a human-readable window into their computation process, and is increasingly relied upon for transparency and…

Human-Computer Interaction · Computer Science 2025-08-29 Mosh Levy , Zohar Elyoseph , Yoav Goldberg

This dissertation investigates how reinforcement learning (RL) methods can be designed to be safe, sample-efficient, and robust. Framed through the unifying perspective of contextual-bandit RL, the work addresses two major application…

Machine Learning · Computer Science 2025-10-20 Shashank Gupta

Obtaining enough labeled data to robustly train complex discriminative models is a major bottleneck in the machine learning pipeline. A popular solution is combining multiple sources of weak supervision using generative models. The…

Machine Learning · Computer Science 2017-09-11 Paroma Varma , Bryan He , Payal Bajaj , Imon Banerjee , Nishith Khandwala , Daniel L. Rubin , Christopher Ré

When inferring reward functions from human behavior (be it demonstrations, comparisons, physical corrections, or e-stops), it has proven useful to model the human as making noisy-rational choices, with a "rationality coefficient" capturing…

Machine Learning · Computer Science 2023-03-10 Gaurav R. Ghosal , Matthew Zurek , Daniel S. Brown , Anca D. Dragan