critique-GRPO
[ICML 2026 Spotlight] Critique-GRPO: Advancing LLM Reasoning with Natural Language and Numerical Feedback
파일 탐색기
최종 버전 다운로드 (.zip)- .DS_Store
- test.parquet
- train.parquet
- run_open_r1_math4k-qwen3-8b-base-critique_simple_gt_online.sh
- dapo_trainer.yaml
- dapo_ray_trainer.py
- main_dapo.py
- prepare_dapo_data.sh
- README.md
- run_dapo_early_qwen2.5_32b.sh
- run_dapo_qwen2.5_32b.sh
- run_dapo_wo_ds_qwen2.5_32b.sh
- test_dapo_7b.sh
- README.md
- prime_trainer.yaml
- __init__.py
- main_prime.py
- prime_core_algos.py
- prime_dp_rm.py
- prime_fsdp_workers.py
- prime_ray_trainer.py
- run_prime_qwen.sh
- evaluation.yaml
- __init__.py
- gpqa.py
- livecodebench.py
- math.py
- __init__.py
- data_process.py
- main_eval.py
- README.md
- reward_score.py
- run_r1_distill_qwen.sh
- sppo_trainer.yaml
- __init__.py
- dp_actor.py
- main_sppo.py
- README.md
- run_qwen2.5-7b_rm.sh
- sppo_ray_trainer.py
- sppo_worker.py
- README.md
- converter_hf_to_mcore.py
- diagnose.py
- install_vllm_sglang_mcore.sh
- model_merger.py
- run_all.sh
- test_tensor_dict.py
- create_dataset.py
- test.parquet
- train.parquet
- config.json
- create_model_tokenizer.py
- generation_config.json
- model.safetensors
- tokenizer_config.json
- main_trainer.py
- README.md
- __init__.py
- task.py
- tokenizer.py
- __init__.py
- run_gen_qwen05.sh
- run_function_reward.sh
- run_model_reward.sh
- run_sft.sh
- test_sp_loss_match.py
- __init__.py
- check_custom_rwd_fn.py
- check_results.py
- run_dapo.sh
- run_gsm8k_fsdp_sgl_multiturn_w_tool.sh
- run_ppo_trainer_megatron.sh
- run_prime.sh
- run_r1_distill_qwen_aime24_eval.sh
- run_ray_trainer.sh
- run_ray_trainer_fire_sampling.sh
- run_ray_trainer_rmpad.sh
- run_sppo.sh
- run_test.sh
- test_memory_buffers.py
- test_ops.py
- test_torch_functional.py
- test_linear_cross_entropy.py
- test_transformer.py
- test_transformers_ulysses.py
- main.py
- test_auto_padding.py
- test_check_worker_alive.py
- test_decorator.py
- test_fused_workers.py
- test_ray_local_envs.py
- client.py
- README.md
- run.sh
- server.py
- test_colocated_workers.py
- test_colocated_workers_fused.py
- test_data_transfer.py
- test_driverfunc_to_worker.py
- test_high_level_scheduling_api.py
- test_rvdz.py
- test_worker_group_basics.py
- test_worker_group_torch.py
- test_sandbox_fusion.py
- test_sandbox.py
- check_license.py
- test_import.py
- test_decorator.py
- __init__.py
- test_metric_utils.py
- __init__.py
- test_fs.py
- test_import_utils.py
- test_model.py
- test_module.py
- test_timeout_decorator.py
- test_fsdp_ckpt.py
- test_multiturn_sft_dataset.py
- test_rl_dataset.py
- test_rm_dataset.py
- test_sft_dataset.py
- test_flops_counter.py
- test_seqlen_balancing.py
- test_torch_functional.py
- async_rollout_utils.py
- run_fsdp_vllm.py
- test_hf_rollout.py
- test_sglang_async_rollout_w_tools.py
- test_sglang_async_spmd.py
- test_sglang_spmd.py
- test_vllm_hf_loader.py
- test_vllm_multi_turn.py
- test_vllm_spmd.py
- test_vllm_tool_calling.py
- utils_sglang.py
- __init__.py
- kill_github_tests.sh
- test_protocol.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- protocol.cpython-310.pyc
- protocol.cpython-311.pyc
- critique_core_algos.cpython-310.pyc
- critique_core_algos.cpython-311.pyc
- critique_dp_actor.cpython-310.pyc
- critique_dp_actor.cpython-311.pyc
- critique_fsdp_workers.cpython-310.pyc
- critique_fsdp_workers.cpython-311.pyc
- critique_main_ppo.cpython-310.pyc
- critique_main_ppo.cpython-311.pyc
- critique_prompts.cpython-310.pyc
- critique_ray_trainer.cpython-310.pyc
- critique_ray_trainer.cpython-311.pyc
- critique_vllm_rollout_spmd.cpython-310.pyc
- critique_vllm_rollout_spmd.cpython-311.pyc
- refinement_prompts.cpython-310.pyc
- rl_dataset_with_critique.cpython-310.pyc
- rl_dataset_with_critique.cpython-311.pyc
- critique_ppo_trainer.yaml
- evaluation.yaml
- generation.yaml
- ppo_megatron_trainer.yaml
- ppo_trainer.yaml
- sft_trainer.yaml
- critique_core_algos.py
- critique_dp_actor.py
- critique_fsdp_workers.py
- critique_main_ppo.py
- critique_prompts.py
- critique_ray_trainer.py
- critique_vllm_rollout_spmd.py
- math_verify_reward.py
- refinement_prompts.py
- reward_with_format.py
- rl_dataset_with_critique.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- registry.cpython-310.pyc
- registry.cpython-311.pyc
- __init__.py
- llama_loader.py
- llama_loader_depracated.py
- llama_saver.py
- __init__.py
- parallel_attention.py
- parallel_decoder.py
- parallel_linear.py
- parallel_mlp.py
- parallel_rmsnorm.py
- __init__.py
- modeling_llama_megatron.py
- __init__.py
- __init__.py
- config_converter.py
- loader.py
- model_forward.py
- model_initializer.py
- readme.md
- registry.py
- saver.py
- util.py
- weight_converter.py
- __init__.py
- qwen2_loader.py
- qwen2_loader_depracated.py
- qwen2_saver.py
- __init__.py
- parallel_attention.py
- parallel_decoder.py
- parallel_linear.py
- parallel_mlp.py
- parallel_rmsnorm.py
- __init__.py
- modeling_qwen2_megatron.py
- __init__.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- llama.cpython-310.pyc
- llama.cpython-311.pyc
- monkey_patch.cpython-310.pyc
- monkey_patch.cpython-311.pyc
- __init__.py
- llama.py
- monkey_patch.py
- qwen2.py
- qwen2_5_vl.py
- qwen2_vl.py
- __init__.py
- README.md
- registry.py
- weight_loader_registry.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- decorator.cpython-310.pyc
- decorator.cpython-311.pyc
- worker.cpython-310.pyc
- worker.cpython-311.pyc
- worker_group.cpython-310.pyc
- worker_group.cpython-311.pyc
- __init__.py
- worker.py
- worker_group.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- ray.cpython-310.pyc
- ray.cpython-311.pyc
- __init__.py
- ray.py
- __init__.py
- decorator.py
- worker.py
- worker_group.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- base.cpython-310.pyc
- base.cpython-311.pyc
- __init__.py
- base.py
- megatron.py
- __init__.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- __init__.py
- parallel_state.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- __init__.py
- arg_utils.py
- config.py
- dtensor_weight_loaders.py
- hf_weight_loader.py
- llm.py
- llm_engine_sp.py
- megatron_weight_loaders.py
- model_loader.py
- model_runner.py
- parallel_state.py
- spmd_gpu_executor.py
- tokenizer.py
- worker.py
- __init__.py
- arg_utils.py
- config.py
- dtensor_weight_loaders.py
- hf_weight_loader.py
- llm.py
- llm_engine_sp.py
- megatron_weight_loaders.py
- model_loader.py
- model_runner.py
- parallel_state.py
- spmd_gpu_executor.py
- tokenizer.py
- worker.py
- __init__.py
- __init__.py
- __init__.py
- base_tool.py
- gsm8k_tool.py
- schemas.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- main_ppo.cpython-310.pyc
- main_ppo.cpython-311.pyc
- critique_ppo_trainer.yaml
- evaluation.yaml
- generation.yaml
- ppo_megatron_trainer.yaml
- ppo_trainer.yaml
- sft_trainer.yaml
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- core_algos.cpython-310.pyc
- core_algos.cpython-311.pyc
- metric_utils.cpython-310.pyc
- metric_utils.cpython-311.pyc
- ray_trainer.cpython-310.pyc
- ray_trainer.cpython-311.pyc
- reward.cpython-310.pyc
- reward.cpython-311.pyc
- __init__.py
- core_algos.py
- metric_utils.py
- ray_trainer.py
- reward.py
- __init__.py
- fsdp_sft_trainer.py
- main_eval.py
- main_generation.py
- main_ppo.py
- runtime_env.yaml
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- flops_counter.cpython-310.pyc
- flops_counter.cpython-311.pyc
- fs.cpython-310.pyc
- fs.cpython-311.pyc
- fsdp_utils.cpython-310.pyc
- fsdp_utils.cpython-311.pyc
- hdfs_io.cpython-310.pyc
- hdfs_io.cpython-311.pyc
- import_utils.cpython-310.pyc
- import_utils.cpython-311.pyc
- logging_utils.cpython-310.pyc
- logging_utils.cpython-311.pyc
- model.cpython-310.pyc
- model.cpython-311.pyc
- py_functional.cpython-310.pyc
- py_functional.cpython-311.pyc
- seqlen_balancing.cpython-310.pyc
- seqlen_balancing.cpython-311.pyc
- tokenizer.cpython-310.pyc
- tokenizer.cpython-311.pyc
- torch_dtypes.cpython-310.pyc
- torch_dtypes.cpython-311.pyc
- torch_functional.cpython-310.pyc
- torch_functional.cpython-311.pyc
- tracking.cpython-310.pyc
- tracking.cpython-311.pyc
- ulysses.cpython-310.pyc
- ulysses.cpython-311.pyc
- vllm_utils.cpython-310.pyc
- vllm_utils.cpython-311.pyc
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- checkpoint_manager.cpython-310.pyc
- checkpoint_manager.cpython-311.pyc
- fsdp_checkpoint_manager.cpython-310.pyc
- fsdp_checkpoint_manager.cpython-311.pyc
- __init__.py
- checkpoint_manager.py
- fsdp_checkpoint_manager.py
- megatron_checkpoint_manager.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- rl_dataset.cpython-310.pyc
- rl_dataset.cpython-311.pyc
- rm_dataset.cpython-310.pyc
- rm_dataset.cpython-311.pyc
- sft_dataset.cpython-310.pyc
- sft_dataset.cpython-311.pyc
- __init__.py
- multiturn_sft_dataset.py
- README.md
- rl_dataset.py
- rm_dataset.py
- sft_dataset.py
- vision_utils.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- performance.cpython-310.pyc
- performance.cpython-311.pyc
- __init__.py
- performance.py
- profile.py
- trajectory_tracker.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- torch_functional.cpython-310.pyc
- torch_functional.cpython-311.pyc
- __init__.py
- torch_functional.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- aggregate_logger.cpython-310.pyc
- aggregate_logger.cpython-311.pyc
- __init__.py
- aggregate_logger.py
- __init__.py
- memory.py
- optimizer.py
- pipeline_parallel.py
- sequence_parallel.py
- tensor_parallel.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- utils.cpython-310.pyc
- utils.cpython-311.pyc
- __init__.py
- utils.py
- __init__.py
- ray_backend.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- hf_math_verify.cpython-310.pyc
- hf_math_verify.cpython-311.pyc
- __init__.py
- testing_util.py
- utils.py
- __init__.py
- grader.py
- math_normalize.py
- examples.cpython-310.pyc
- examples.cpython-311.pyc
- examples.cpython-39.pyc
- parser.cpython-310.pyc
- parser.cpython-311.pyc
- parser.cpython-39.pyc
- utils.cpython-310.pyc
- utils.cpython-311.pyc
- utils.cpython-39.pyc
- examples.py
- grader.py
- parser.py
- utils.py
- __init__.py
- utils.py
- __init__.py
- geo3k.py
- gsm8k.py
- hf_math_verify.py
- math.py
- math_batch.py
- math_dapo.py
- math_verify.py
- __init__.py
- config.py
- distributed.py
- flops_counter.py
- fs.py
- fsdp_utils.py
- hdfs_io.py
- import_utils.py
- logging_utils.py
- megatron_utils.py
- memory_buffer.py
- model.py
- net_utils.py
- py_functional.py
- ray_utils.py
- seqlen_balancing.py
- tokenizer.py
- torch_dtypes.py
- torch_functional.py
- tracking.py
- ulysses.py
- vllm_utils.py
- version
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- base.cpython-310.pyc
- base.cpython-311.pyc
- dp_actor.cpython-310.pyc
- dp_actor.cpython-311.pyc
- __init__.py
- base.py
- dp_actor.py
- megatron_actor.py
- __init__.py
- base.py
- dp_critic.py
- megatron_critic.py
- __init__.py
- batch.py
- dapo.py
- naive.py
- prime.py
- __init__.py
- reward_model.py
- __init__.py
- base.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- async_server.cpython-310.pyc
- async_server.cpython-311.pyc
- base.cpython-310.pyc
- base.cpython-311.pyc
- hf_rollout.cpython-310.pyc
- hf_rollout.cpython-311.pyc
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- naive_rollout.cpython-310.pyc
- naive_rollout.cpython-311.pyc
- __init__.py
- naive_rollout.py
- __init__.py
- async_sglang_rollout.py
- sglang_rollout.py
- utils.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- vllm_rollout_spmd.cpython-310.pyc
- vllm_rollout_spmd.cpython-311.pyc
- __init__.py
- fire_vllm_rollout.py
- vllm_async_server.py
- vllm_rollout.py
- vllm_rollout_spmd.py
- __init__.py
- async_server.py
- base.py
- hf_rollout.py
- schemas.py
- tokenizer.py
- __init__.cpython-310.pyc
- __init__.cpython-311.pyc
- base.cpython-310.pyc
- base.cpython-311.pyc
- fsdp_ulysses.cpython-310.pyc
- fsdp_ulysses.cpython-311.pyc
- fsdp_vllm.cpython-310.pyc
- fsdp_vllm.cpython-311.pyc
- __init__.py
- base.py
- fsdp_sglang.py
- fsdp_ulysses.py
- fsdp_vllm.py
- megatron_sglang.py
- megatron_vllm.py
- __init__.py
- fsdp_workers.py
- megatron_workers.py
- .DS_Store
- __init__.py
- protocol.py
- .DS_Store
- .pre-commit-config.yaml
- .readthedocs.yaml
- LICENSE
- Notice.txt
- pyproject.toml
- README.md
- requirements.txt
- requirements_sglang.txt
- setup.py
- .DS_Store
- Critique_GRPO.png
- Introduction.png
- README.md
- Three_types_of_critique.png
- training_env.yml
// repository documentation
Was this content helpful?
(0 ratings)
