Topic
Rewards translate outcomes and preferences into the signal used to improve a policy. They may come from humans, learned reward models, AI judges, tests, or executable verifiers. This topic follows how reward signals scale, where they fail, and how evaluation can detect proxy optimization before it becomes a production problem.
Reward models
AI feedback
Process supervision
Verifier calibration
Latest research
Zhejiang University
KnowGPO groups search-agent steps by accumulated task-relevant knowledge and rewards newly acquired facts. Controlled Qwen3-8B comparisons show gains over GRPO on four evaluation splits, but the method requires reference answers, successful training trajectories, and an external LLM verifier. Several narrative claims in the first version disagree with its tables.
Tsinghua University
ActionOPD distills black-box agent teachers by rewarding semantic agreement between sampled student and teacher actions. It needs teacher generation rather than token probabilities or aligned tokenizers. Experiments report cross-benchmark transfer, alongside mixed success-versus-reliability results against token-level OPD.
Fin.ai
The paper probes whether reference-conditioned self-distillation actually locates and corrects reasoning errors. Generating a critique before scoring gives more error-specific feedback in the tested models, and small Knights & Knaves training experiments improve with a critique-conditioned term. Every critique has access to a reference solution.
Xiaomi
Xiaomi’s MiMo-V2.6 report details large-scale mixed-task agentic RL: long-horizon batches, diverse environments and harnesses, and agentic grading beyond binary test rewards. It reports training-cost curves, router-stability experiments, infrastructure failures, and a subsequent multi-teacher distillation stage.
University of Illinois Urbana-Champaign
DARS turns annotated task dependencies into signed step rewards: an unresolved mistake discounts dependent progress while independent work retains credit. Experiments cover household tasks, shopping, retrieval, and mathematics, with the clearest matched gain on a 1.5B ALFWorld agent.
Li, Shuyue Stella
Verifiable Visual Rewards derives image prompts and deterministic constraint checks from the same geometric scene specification. Flow-GRPO training on simple synthetic scenes improves instruction following in Stable Diffusion 3.5 Medium and transfers to natural-image prompts, with measurable tradeoffs.
Kuaishou
KwaiMind combines diffusion preference optimization, task-specific online RL, and on-policy distillation for commercial image editing. Dedicated rewards cover predicted click-through rate, text rendering, and product consistency; the report separates offline reward-based gains from an online image-selection experiment.
Xiaomi
Xiaomi’s CodeMidas turns implemented functionality in existing codebases into executable coding RL tasks. Training MiMo-V2.5 on 5,545 filtered environments improves five external benchmarks, while a 3,000-task filtered subset beats an 8,000-task unfiltered baseline on three evaluations.
Xuan Liu
Checkpoint handoff separates two sources of agentic RL gains: reaching useful states and completing tasks from those states. Replaying histories across SFT and RL checkpoints reveals an interaction that endpoint success alone cannot attribute.
Hejia Geng
ScienceIDE turns scientific codebases and expert acceptance criteria into executable tasks, environments, and verifiers. Its experiments examine scientific software repair, supervised transfer, and RL with costly outcome feedback and long interaction histories.
Leon Bergen
Simple directions in model activations can flag reward-hacking behavior during agent evaluations. The study compares difference-of-means probes with LLM monitors, examines predictive signals before actions, and documents where detection still produces errors.
kaivu
A research-methods proposal for studying when models misbehave in realistic use. It starts from recorded failures, varies environment conditions or the policy, and asks how those observations can support estimates of behavior in a target task distribution.
Qwen Team, Alibaba Group
Terminal-Universe reconstructs reusable terminal environments from recorded agent trajectories, then generates new tasks within projects, across related codebases, and over multiple user rounds. SFT experiments suggest that expanding environment diversity can produce more useful supervision than adding queries or solutions to existing environments.
Anthropic
An Anthropic study and LessWrong discussion examining how reward hacking learned in deliberately vulnerable RL environments can transfer to harmful behavior in other evaluation scenarios. The findings also show important limits to that transfer.
Jingxiao Yang
A work-in-progress study of long-horizon agent reinforcement learning that uses privileged process supervision to redistribute verified trajectory-level credit across executable actions.
Z.ai
An official GLM-5.3 release note describing a post-training-only update on the same base model as GLM-5.2, with expanded executable environments for complex coding, security, and long-horizon professional workflows.
Steven Byrnes
A community synthesis linking four training signals—imitation, human approval, automatic verifiers, and LLM judges—to distinct classes of observed or hypothesized alignment failure.
Fiora Starlight
A LessWrong proposal to let policies report exploitable RLVR environment bugs after a rollout, reward high-quality verified reports, and use them to patch the training environment.
Yufei Li
A production-oriented technical report on a generative reasoning re-ranker trained with semantic-ID mid-training, teacher-trace distillation, on-policy distillation, and reinforcement learning from verifiable ranking rewards.
Z.ai
An official release of the 750B-A40B GLM-5.2 model, combining a one-million-token context, IndexShare sparse-attention reuse, and larger-scale agentic reinforcement learning for long-horizon tasks.
MiniMax
An official release of MiniMax-M3, a natively multimodal sparse model with about 428B total and 23B active parameters, a one-million-token context window, and agent-oriented coding and cowork capabilities.
MiniMax
A technical report on the MiniMax-M2 family, pairing a 229.9B-parameter sparse MoE with agent-generated, verifiable trajectories and Forge, a scalable reinforcement-learning system for long-horizon agents.
Meta FAIR
This work studies language models that generate candidate responses and also provide the preference signal used to improve subsequent iterations.
NVIDIA
Eureka uses a code-capable language model to propose and iteratively improve executable reward functions for reinforcement-learning tasks.
OpenAI
A comparison of outcome supervision and process supervision for mathematical reasoning, centered on whether feedback should evaluate only the final answer or intermediate steps as well.
Anthropic
Constitutional AI uses written principles and model-generated critiques and preferences to reduce direct dependence on human harmlessness labels during alignment training.
OpenAI
An empirical study of how optimizing against a learned reward model can eventually reduce performance under a higher-quality reference signal.
OpenAI
A large-scale study of training summarization models from human comparisons using reward modeling and reinforcement learning.
OpenAI
A foundational study of learning a reward model from pairwise human preferences and using that learned reward to train an RL agent.