Related papers: Detecting and Suppressing Reward Hacking with Grad…

Reinforcement learning with verifiable outcome rewards (RLVR) has effectively scaled up chain-of-thought (CoT) reasoning in large language models (LLMs). Yet, its efficacy in training vision-language model (VLM) agents for goal-directed…

Models capable of "thinking with images" by dynamically grounding their reasoning in visual evidence represent a major leap in multimodal AI. However, replicating and advancing this ability is non-trivial, with current methods often trapped…

Reinforcement Learning with Verifiable Rewards (RLVR) improves final-answer accuracy on reasoning tasks, but it does not reliably improve reasoning quality. Because outcome rewards only assess final answers, they also reward spurious…

The prevailing paradigm for training large reasoning models--combining Supervised Fine-Tuning (SFT) with Reinforcement Learning with Verifiable Rewards (RLVR)--is fundamentally constrained by its reliance on high-quality, human-annotated…

Recent advancements in the Generative Reward Model (GRM) have demonstrated its potential to enhance the reasoning abilities of LLMs through Chain-of-Thought (CoT) prompting. Despite these gains, existing implementations of GRM suffer from…

Reinforcement learning (RL) systems typically optimize scalar reward functions that assume precise and reliable evaluation of outcomes. However, real-world objectives--especially those derived from human preferences--are often uncertain,…

Reinforcement Learning with Verifiable Rewards (RLVR) has catalyzed significant breakthroughs in complex LLM reasoning within verifiable domains, such as mathematics and programming. Recent efforts have sought to extend this paradigm to…

Looped Language Models (LoopLMs) perform multi-step latent reasoning prior to token generation and outperform conventional LLMs on reasoning benchmarks at smaller parameter budgets. However, attempts to further improve LoopLM reasoning with…

As large language models become smaller and more efficient, small reasoning models (SRMs) are crucial for enabling chain-of-thought (CoT) reasoning in resource-constrained settings. However, they are prone to faithfulness hallucinations,…

Red teaming is a common strategy for identifying weaknesses in generative language models (LMs), where adversarial prompts are produced that trigger an LM to generate unsafe responses. Red teaming is instrumental for both model alignment…

Aligning generative recommender systems to user preferences via post-training is critical for closing the gap between next-item prediction and actual recommendation quality. Existing post-training methods are ill-suited for production-scale…

Reinforcement Learning (RL) serves as a potent paradigm for enhancing reasoning capabilities in Large Language Models (LLMs), yet standard outcome-based approaches often suffer from reward sparsity and inefficient credit assignment. In this…

Reinforcement Learning from Human Feedback (RLHF) enables powerful LLM alignment but can introduce reward hacking - models exploit spurious correlations in proxy rewards without genuine alignment. Compounding this, the objectives…

Reward hacking in code generation, where models exploit evaluation loopholes to obtain full reward without correctly solving the tasks, poses a critical challenge for Reinforcement Learning (RL) and the deployment of reasoning models.…

In practice, rigorous reasoning is often a key driver of correct code, while Reinforcement Learning (RL) for code generation often neglects optimizing reasoning quality. Bringing process-level supervision into RL is appealing, but it faces…

Generative Reward Models (GRMs) provide greater flexibility than scalar reward models in capturing human preferences, but their effectiveness is limited by poor reasoning capabilities. This often results in incomplete or overly speculative…

Reinforcement learning with verifiable rewards (RLVR) has become a powerful paradigm for improving language models on reasoning-intensive tasks, but its effectiveness is often limited by exploration. For example, models often fail on hard…

In financial Retrieval-Augmented Generation (RAG) systems, models frequently rely on retrieved documents to generate accurate responses due to the time-sensitive nature of the financial domain. While retrieved documents help address…

State-of-the-art reasoning models utilize long chain-of-thought (CoT) to solve increasingly complex problems using more test-time computation. In this work, we explore a long CoT setting where the model makes up to K successive attempts at…

The growing disparity between the exponential scaling of computational resources and the finite growth of high-quality text data now constrains conventional scaling approaches for large language models (LLMs). To address this challenge, we…