Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2603. 19329v3 Announce Type: replace-cross Abstract: Large language models (LLMs) can generate plausible code but offer limited guarantees of correctness.
arXiv:2605. 24396v2 Announce Type: replace Abstract: Long chains of thought (CoT) from current language models frequently contain logical gaps and unjustified leaps, limiting the gains from additional test-time compute.
arXiv:2506. 07223v2 Announce Type: replace Abstract: Large language models (LLMs) have substantially improved the planning capabilities of embodied agents, enabling their deployment in dynamic and safety-critical environments.
arXiv:2608. 07719v1 Announce Type: new Abstract: Offline reinforcement learning repeatedly trains policies from a fixed transition pool, making redundant data costly across seeds and hyperparameters, while naive subsampling can remove rare transitions needed for long-horizon credit assignment.
arXiv:2608. 07725v1 Announce Type: new Abstract: Average-reward reinforcement-learning regret is known up to logarithmic factors, but the numerical content of published guarantees is difficult to compare because probability mode, structural parameter, logarithmic normalization, prior information, and planning assumptions differ.
arXiv:2508. 03875v2 Announce Type: replace Abstract: Many sequential decision problems offer qualitatively different ways of influencing the environment: some interventions act immediately, whereas others induce persistent effects that continue to shape future states long after the decision that initiated them.
arXiv:2608. 07527v1 Announce Type: cross Abstract: Long-document understanding requires models to find and combine evidence across many pages, layouts, tables, figures, and charts.
arXiv:2608. 09128v1 Announce Type: cross Abstract: LLM agents are increasingly deployed in multi-agent social settings where they must cooperate, negotiate, and adapt to other agents.
arXiv:2608. 09217v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become a central post-training paradigm for eliciting reasoning capabilities in large language models, yet uniform task sampling allocates compute without regard to differences in how tasks respond to optimization.
arXiv:2608. 08255v1 Announce Type: new Abstract: Agentic reinforcement learning (RL) often suffers from delayed and sparse rewards in real-world environments.
arXiv:2404. 03082v3 Announce Type: replace Abstract: Partially ordered sets (posets) are discrete mathematical structures that formalize the notion of comparison without forcing every pair of objects to be comparable.
arXiv:2604. 19569v5 Announce Type: replace-cross Abstract: Q-learning is a fundamental algorithmic primitive in reinforcement learning.
arXiv:2608. 09617v1 Announce Type: new Abstract: Symbolic regression is the problem of finding an algebraic expression describing a stochastic dependence of a target variable on a set of inputs.
arXiv:2511. 10453v4 Announce Type: replace-cross Abstract: Large language models often respond to ambiguous requests by implicitly committing to one interpretation, frustrating users and creating safety risks when that interpretation is wrong.
arXiv:2608. 08878v1 Announce Type: cross Abstract: Transformer-based large language models (LLMs) achieve strong performance across many tasks, but their Key-Value (KV) cache grows linearly with sequence length, creating a severe memory bottleneck for long-context inference.
arXiv:2608. 09826v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards yields no group-relative signal when rollout groups are uniformly correct or uniformly wrong, which account for 63.
Large language models (LLMs) have advanced code generation, where executable feedback provides a more reliable learning signal than textual imitation alone. Hardware verification is an important application of code generation and accounts for a substantial fraction of modern chip design effort, with high-coverage testbench stimulus generation as a key task.
We introduce the Dark Souls Learning Environment (DSLE), a containerized platform that presents all 22 boss encounters of Dark Souls: Remastered as game-playing agent benchmarks through a Gymnasium-style interface. DSLE combines real-time combat, high-dimensional visual input, and sparse terminal rewards, with each environment step being a real action executed against the running game.
Reinforcement learning with verifiable rewards yields no group-relative signal when rollout groups are uniformly correct or uniformly wrong, which account for 63. 0-68.
Diffusion models have enabled high-quality video generation in recent years, but the high cost of iterative sampling hinders their practical deployment. Few-step distillation alleviates this cost, yet exposes a quality--diversity trade-off between its two dominant paradigms: trajectory-level distillation (e.