Official Project Page for KL-Regularized Policy Optimization (KLPO) (https://arxiv.org/abs/2610.08963)
☆191Nov 5, 2026Updated this week
Alternatives and similar repositories for KLPO
Users that are interested in KLPO are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- FlashREINFORCE: Critic-Free, Single-Rollout, Asynchronous RL for Agentic Language Models☆106Sep 14, 2026Updated last month
- Official Project Page for Recurrent Looped Transformer (RLT) (https://arxiv.org/abs/2610.07591)☆1,140Updated this week
- Laws about stateful rules, proven in TypeScript: a small trusted kernel, an untrusted prover, a Lean twin, and GPU searches for counterex…☆29Updated this week
- TEVO: evolve LM motifs cheaply, then validate them in downstream train.py loops.☆20Apr 18, 2026Updated 6 months ago
- ☆13Aug 27, 2026Updated 2 months ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- A scalable, agentic-first, and HuggingFace-native RL framework for research (9k lines).☆1,218Updated this week
- ☆56Oct 20, 2026Updated 2 weeks ago
- Learning to attribution language model outputs to representations and weights☆41Sep 25, 2026Updated last month
- JEPA-Anything: Learning Predictive Models across Different Worlds☆397Sep 18, 2026Updated last month
- PyTorch code for our paper "Progressive Binarization with Semi-Structured Pruning for LLMs"☆12Jul 11, 2026Updated 3 months ago
- [CVPR2026] Mistake Attribution: Fine-Grained Mistake Understanding in Egocentric Videos☆16Aug 24, 2026Updated 2 months ago
- [CVPR 2025 Highlight] ESC: Erasing Space Concept for Knowledge Deletion☆12Dec 8, 2025Updated 10 months ago
- Flow Map Speculative Decoding☆56Updated this week
- ☆13Sep 29, 2026Updated last month
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- ☆1,368Sep 16, 2026Updated last month
- GLOSS: Geometric Local Self-Similarity Learning for Faithful Reference-Guided Texture Fill (SIGGRAPH Asia 2026)☆19Oct 6, 2026Updated last month
- An open protocol for distributed multi-agent autoformalization.☆118Oct 18, 2026Updated 2 weeks ago
- RSI-Anything: Unlocking Recursive Self-Improvement with World Models☆1,126Oct 13, 2026Updated 3 weeks ago
- Code for Latent Sparse Attention: Towards Subquadratic Ultra-Long Context Attention☆421Updated this week
- Official Project Page for Falcon: Fast Weight Attention for Continual Learning (https://arxiv.org/abs/2608.27763)☆41Updated this week
- ☆20Updated this week
- OPAL: a reactive optimizer. A coding agent rewrites the optimizer while your model trains.☆354Updated this week
- ☆131Updated this week
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- CUDA Blackwell performance fork of antirez/ds4 (DeepSeek-V4-Flash): ~4x prefill, ~1.5x decode, continuous batching, weight server, DSpark☆77Oct 26, 2026Updated last week
- Halo is an open-source framework built by White Circle for training large language and multimodal models☆847Updated this week
- Speedrun for pretraining without backpropagation☆450Updated this week
- Train Embedding Models on MLX.☆17Jun 2, 2026Updated 5 months ago
- Documentation pages for web ontologies.☆78Updated this week
- ☆11Jul 20, 2026Updated 3 months ago
- Reason over explicit states, many branches at once: merge duplicates, cancel dead ends, advance together.☆181Oct 3, 2026Updated last month
- Official codebase for "Multi-Token Prediction Done Right"☆94Oct 30, 2026Updated last week
- Code for Scalable Offline Model-Based RL with Action chunking☆36Feb 20, 2026Updated 8 months ago
- End-to-end encrypted email - Proton Mail • AdSpecial offer: 40% Off Yearly / 80% Off First Month. All Proton services are open source and independently audited for security.
- A very-low-resource API-compatible replacement for Ray☆13Updated this week
- Experiments for the spectral neuron paper☆23Sep 5, 2026Updated 2 months ago
- A self-hosted drop-in replacement for TypeSafe's jev, powered by GliFormer.☆312Sep 20, 2026Updated last month
- Cuda kernels for leveraging LLM sparsity to improve throughput and decrease the memory requirements during inference and training.☆262Jun 29, 2026Updated 4 months ago
- Linux anonymity / OPSEC toolkit — Tor-by-default with a kill switch, leak verification, and pluggable upstreams. Authorized use only.☆24Sep 2, 2026Updated 2 months ago
- Llama-architecture models (SmolLM2 135M, 360M, 1.7B) in Bend 2, token for token against PyTorch, with the shapes in the types☆16Updated this week
- Bend2 - WIP☆16Jun 17, 2025Updated last year