├── .gitignore ├── README.md ├── RL ├── .pre-commit-config.yaml ├── Dockerfile ├── Dockerfile.legacy ├── LICENSE ├── Makefile ├── base.text ├── download.sh ├── examples │ ├── config_monet.yaml │ ├── dataset_valid_ids │ │ ├── Geometry3K │ │ │ └── valid_ids.txt │ │ ├── Thyme-train │ │ │ └── valid_ids.txt │ │ └── Thyme-val │ │ │ └── valid_ids.txt │ ├── format_prompt │ │ └── monet_format.jinja │ ├── merge_model.sh │ ├── reward_function │ │ ├── answer_transformation.py │ │ ├── eval_grader.py │ │ ├── monet_reward_function.py │ │ └── r1v.py │ ├── runtime_env.yaml │ └── vlpo_train.sh ├── monet_models │ ├── transformers │ │ └── monet_modeling_qwen2_5_vl.py │ └── vllm │ │ ├── __init__.py │ │ ├── latent_hook.py │ │ ├── latent_recorder.py │ │ └── monet_gpu_model_runner.py ├── monet_rl_patch.py ├── pyproject.toml ├── ray_tmp │ ├── session_2025-10-26_20-19-12_841302_26777 │ │ ├── node_ip_address.json │ │ └── node_ip_address.json.lock │ ├── session_2025-10-26_20-21-49_985342_26922 │ │ ├── node_ip_address.json │ │ └── node_ip_address.json.lock │ └── session_latest ├── requirements.txt ├── setup.py ├── tools │ ├── actors.py │ ├── api_judge.py │ ├── custom_api.py │ └── hash_dict.py └── verl │ ├── __init__.py │ ├── models │ ├── __init__.py │ ├── monkey_patch.py │ └── transformers │ │ ├── __init__.py │ │ ├── flash_attention_utils.py │ │ └── qwen2_vl.py │ ├── protocol.py │ ├── single_controller │ ├── __init__.py │ ├── base │ │ ├── __init__.py │ │ ├── decorator.py │ │ ├── register_center │ │ │ ├── __init__.py │ │ │ └── ray.py │ │ ├── worker.py │ │ └── worker_group.py │ └── ray │ │ ├── __init__.py │ │ └── base.py │ ├── trainer │ ├── EasyR1 │ │ └── verl │ │ │ └── workers │ │ │ └── fsdp_workers.py │ ├── __init__.py │ ├── config.py │ ├── core_algos.py │ ├── data_loader.py │ ├── main.py │ ├── metrics.py │ ├── ray_trainer.py │ └── save_any_log.py │ ├── utils │ ├── __init__.py │ ├── checkpoint │ │ ├── __init__.py │ │ ├── checkpoint_manager.py │ │ └── fsdp_checkpoint_manager.py │ ├── dataset.py │ ├── flops_counter.py │ ├── fsdp_utils.py │ ├── logger │ │ ├── __init__.py │ │ ├── gen_logger.py │ │ └── logger.py │ ├── model_utils.py │ ├── py_functional.py │ ├── seqlen_balancing.py │ ├── tokenizer.py │ ├── torch_dtypes.py │ ├── torch_functional.py │ └── ulysses.py │ └── workers │ ├── __init__.py │ ├── actor │ ├── __init__.py │ ├── base.py │ ├── config.py │ ├── dp_actor.py │ └── fa_shim.py │ ├── config.py │ ├── critic │ ├── __init__.py │ ├── base.py │ ├── config.py │ └── dp_critic.py │ ├── fsdp_workers.py │ ├── reward │ ├── __init__.py │ ├── config.py │ └── function.py │ ├── rollout │ ├── __init__.py │ ├── base.py │ ├── config.py │ ├── utils │ │ ├── checker.py │ │ ├── math_equal.py │ │ ├── util.py │ │ ├── xwin_latex_answer_check.py │ │ └── xwin_parsing_lib.py │ └── vllm_rollout_spmd.py │ └── sharding_manager │ ├── __init__.py │ ├── base.py │ ├── fsdp_ulysses.py │ └── fsdp_vllm.py ├── deepspeed └── ds_zero2_gpu.json ├── images ├── example_question.png └── overview.png ├── inference ├── apply_vllm_monet.py ├── example.sh ├── load_and_gen_vllm.py ├── vllm │ └── monet_gpu_model_runner.py └── vllm_inference_example.py ├── monet_qwen_model ├── apply_qwen2_5_monet.py └── modeling_qwen2_5_vl_monet.py ├── requirements.txt ├── script_examples ├── sft_stage1.sh ├── sft_stage2.sh └── sft_stage3.sh ├── src ├── main.py ├── precompute_teacher_latents.py ├── precompute_teacher_reps.py ├── task.py ├── trainer.py └── utils.py └── transformers └── .DS_Store /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/.gitignore -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/README.md -------------------------------------------------------------------------------- /RL/.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/.pre-commit-config.yaml -------------------------------------------------------------------------------- /RL/Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/Dockerfile -------------------------------------------------------------------------------- /RL/Dockerfile.legacy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/Dockerfile.legacy -------------------------------------------------------------------------------- /RL/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/LICENSE -------------------------------------------------------------------------------- /RL/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/Makefile -------------------------------------------------------------------------------- /RL/base.text: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/base.text -------------------------------------------------------------------------------- /RL/download.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/download.sh -------------------------------------------------------------------------------- /RL/examples/config_monet.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/config_monet.yaml -------------------------------------------------------------------------------- /RL/examples/dataset_valid_ids/Geometry3K/valid_ids.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/dataset_valid_ids/Geometry3K/valid_ids.txt -------------------------------------------------------------------------------- /RL/examples/dataset_valid_ids/Thyme-train/valid_ids.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/dataset_valid_ids/Thyme-train/valid_ids.txt -------------------------------------------------------------------------------- /RL/examples/dataset_valid_ids/Thyme-val/valid_ids.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/dataset_valid_ids/Thyme-val/valid_ids.txt -------------------------------------------------------------------------------- /RL/examples/format_prompt/monet_format.jinja: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/format_prompt/monet_format.jinja -------------------------------------------------------------------------------- /RL/examples/merge_model.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/merge_model.sh -------------------------------------------------------------------------------- /RL/examples/reward_function/answer_transformation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/reward_function/answer_transformation.py -------------------------------------------------------------------------------- /RL/examples/reward_function/eval_grader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/reward_function/eval_grader.py -------------------------------------------------------------------------------- /RL/examples/reward_function/monet_reward_function.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/reward_function/monet_reward_function.py -------------------------------------------------------------------------------- /RL/examples/reward_function/r1v.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/reward_function/r1v.py -------------------------------------------------------------------------------- /RL/examples/runtime_env.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/runtime_env.yaml -------------------------------------------------------------------------------- /RL/examples/vlpo_train.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/examples/vlpo_train.sh -------------------------------------------------------------------------------- /RL/monet_models/transformers/monet_modeling_qwen2_5_vl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/monet_models/transformers/monet_modeling_qwen2_5_vl.py -------------------------------------------------------------------------------- /RL/monet_models/vllm/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/monet_models/vllm/__init__.py -------------------------------------------------------------------------------- /RL/monet_models/vllm/latent_hook.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/monet_models/vllm/latent_hook.py -------------------------------------------------------------------------------- /RL/monet_models/vllm/latent_recorder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/monet_models/vllm/latent_recorder.py -------------------------------------------------------------------------------- /RL/monet_models/vllm/monet_gpu_model_runner.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/monet_models/vllm/monet_gpu_model_runner.py -------------------------------------------------------------------------------- /RL/monet_rl_patch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/monet_rl_patch.py -------------------------------------------------------------------------------- /RL/pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/pyproject.toml -------------------------------------------------------------------------------- /RL/ray_tmp/session_2025-10-26_20-19-12_841302_26777/node_ip_address.json: -------------------------------------------------------------------------------- 1 | {"node_ip_address": "10.48.94.28"} -------------------------------------------------------------------------------- /RL/ray_tmp/session_2025-10-26_20-19-12_841302_26777/node_ip_address.json.lock: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /RL/ray_tmp/session_2025-10-26_20-21-49_985342_26922/node_ip_address.json: -------------------------------------------------------------------------------- 1 | {"node_ip_address": "10.48.94.28"} -------------------------------------------------------------------------------- /RL/ray_tmp/session_2025-10-26_20-21-49_985342_26922/node_ip_address.json.lock: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /RL/ray_tmp/session_latest: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/ray_tmp/session_latest -------------------------------------------------------------------------------- /RL/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/requirements.txt -------------------------------------------------------------------------------- /RL/setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/setup.py -------------------------------------------------------------------------------- /RL/tools/actors.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/tools/actors.py -------------------------------------------------------------------------------- /RL/tools/api_judge.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/tools/api_judge.py -------------------------------------------------------------------------------- /RL/tools/custom_api.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/tools/custom_api.py -------------------------------------------------------------------------------- /RL/tools/hash_dict.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/tools/hash_dict.py -------------------------------------------------------------------------------- /RL/verl/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/__init__.py -------------------------------------------------------------------------------- /RL/verl/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/models/__init__.py -------------------------------------------------------------------------------- /RL/verl/models/monkey_patch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/models/monkey_patch.py -------------------------------------------------------------------------------- /RL/verl/models/transformers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/models/transformers/__init__.py -------------------------------------------------------------------------------- /RL/verl/models/transformers/flash_attention_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/models/transformers/flash_attention_utils.py -------------------------------------------------------------------------------- /RL/verl/models/transformers/qwen2_vl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/models/transformers/qwen2_vl.py -------------------------------------------------------------------------------- /RL/verl/protocol.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/protocol.py -------------------------------------------------------------------------------- /RL/verl/single_controller/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/__init__.py -------------------------------------------------------------------------------- /RL/verl/single_controller/base/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/base/__init__.py -------------------------------------------------------------------------------- /RL/verl/single_controller/base/decorator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/base/decorator.py -------------------------------------------------------------------------------- /RL/verl/single_controller/base/register_center/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/base/register_center/__init__.py -------------------------------------------------------------------------------- /RL/verl/single_controller/base/register_center/ray.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/base/register_center/ray.py -------------------------------------------------------------------------------- /RL/verl/single_controller/base/worker.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/base/worker.py -------------------------------------------------------------------------------- /RL/verl/single_controller/base/worker_group.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/base/worker_group.py -------------------------------------------------------------------------------- /RL/verl/single_controller/ray/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/ray/__init__.py -------------------------------------------------------------------------------- /RL/verl/single_controller/ray/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/single_controller/ray/base.py -------------------------------------------------------------------------------- /RL/verl/trainer/EasyR1/verl/workers/fsdp_workers.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /RL/verl/trainer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/__init__.py -------------------------------------------------------------------------------- /RL/verl/trainer/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/config.py -------------------------------------------------------------------------------- /RL/verl/trainer/core_algos.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/core_algos.py -------------------------------------------------------------------------------- /RL/verl/trainer/data_loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/data_loader.py -------------------------------------------------------------------------------- /RL/verl/trainer/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/main.py -------------------------------------------------------------------------------- /RL/verl/trainer/metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/metrics.py -------------------------------------------------------------------------------- /RL/verl/trainer/ray_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/ray_trainer.py -------------------------------------------------------------------------------- /RL/verl/trainer/save_any_log.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/trainer/save_any_log.py -------------------------------------------------------------------------------- /RL/verl/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/__init__.py -------------------------------------------------------------------------------- /RL/verl/utils/checkpoint/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/checkpoint/__init__.py -------------------------------------------------------------------------------- /RL/verl/utils/checkpoint/checkpoint_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/checkpoint/checkpoint_manager.py -------------------------------------------------------------------------------- /RL/verl/utils/checkpoint/fsdp_checkpoint_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/checkpoint/fsdp_checkpoint_manager.py -------------------------------------------------------------------------------- /RL/verl/utils/dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/dataset.py -------------------------------------------------------------------------------- /RL/verl/utils/flops_counter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/flops_counter.py -------------------------------------------------------------------------------- /RL/verl/utils/fsdp_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/fsdp_utils.py -------------------------------------------------------------------------------- /RL/verl/utils/logger/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/logger/__init__.py -------------------------------------------------------------------------------- /RL/verl/utils/logger/gen_logger.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/logger/gen_logger.py -------------------------------------------------------------------------------- /RL/verl/utils/logger/logger.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/logger/logger.py -------------------------------------------------------------------------------- /RL/verl/utils/model_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/model_utils.py -------------------------------------------------------------------------------- /RL/verl/utils/py_functional.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/py_functional.py -------------------------------------------------------------------------------- /RL/verl/utils/seqlen_balancing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/seqlen_balancing.py -------------------------------------------------------------------------------- /RL/verl/utils/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/tokenizer.py -------------------------------------------------------------------------------- /RL/verl/utils/torch_dtypes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/torch_dtypes.py -------------------------------------------------------------------------------- /RL/verl/utils/torch_functional.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/torch_functional.py -------------------------------------------------------------------------------- /RL/verl/utils/ulysses.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/utils/ulysses.py -------------------------------------------------------------------------------- /RL/verl/workers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/__init__.py -------------------------------------------------------------------------------- /RL/verl/workers/actor/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/actor/__init__.py -------------------------------------------------------------------------------- /RL/verl/workers/actor/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/actor/base.py -------------------------------------------------------------------------------- /RL/verl/workers/actor/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/actor/config.py -------------------------------------------------------------------------------- /RL/verl/workers/actor/dp_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/actor/dp_actor.py -------------------------------------------------------------------------------- /RL/verl/workers/actor/fa_shim.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/actor/fa_shim.py -------------------------------------------------------------------------------- /RL/verl/workers/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/config.py -------------------------------------------------------------------------------- /RL/verl/workers/critic/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/critic/__init__.py -------------------------------------------------------------------------------- /RL/verl/workers/critic/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/critic/base.py -------------------------------------------------------------------------------- /RL/verl/workers/critic/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/critic/config.py -------------------------------------------------------------------------------- /RL/verl/workers/critic/dp_critic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/critic/dp_critic.py -------------------------------------------------------------------------------- /RL/verl/workers/fsdp_workers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/fsdp_workers.py -------------------------------------------------------------------------------- /RL/verl/workers/reward/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/reward/__init__.py -------------------------------------------------------------------------------- /RL/verl/workers/reward/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/reward/config.py -------------------------------------------------------------------------------- /RL/verl/workers/reward/function.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/reward/function.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/__init__.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/base.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/config.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/utils/checker.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/utils/checker.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/utils/math_equal.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/utils/math_equal.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/utils/util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/utils/util.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/utils/xwin_latex_answer_check.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/utils/xwin_latex_answer_check.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/utils/xwin_parsing_lib.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/utils/xwin_parsing_lib.py -------------------------------------------------------------------------------- /RL/verl/workers/rollout/vllm_rollout_spmd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/rollout/vllm_rollout_spmd.py -------------------------------------------------------------------------------- /RL/verl/workers/sharding_manager/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/sharding_manager/__init__.py -------------------------------------------------------------------------------- /RL/verl/workers/sharding_manager/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/sharding_manager/base.py -------------------------------------------------------------------------------- /RL/verl/workers/sharding_manager/fsdp_ulysses.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/sharding_manager/fsdp_ulysses.py -------------------------------------------------------------------------------- /RL/verl/workers/sharding_manager/fsdp_vllm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/RL/verl/workers/sharding_manager/fsdp_vllm.py -------------------------------------------------------------------------------- /deepspeed/ds_zero2_gpu.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/deepspeed/ds_zero2_gpu.json -------------------------------------------------------------------------------- /images/example_question.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/images/example_question.png -------------------------------------------------------------------------------- /images/overview.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/images/overview.png -------------------------------------------------------------------------------- /inference/apply_vllm_monet.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/inference/apply_vllm_monet.py -------------------------------------------------------------------------------- /inference/example.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/inference/example.sh -------------------------------------------------------------------------------- /inference/load_and_gen_vllm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/inference/load_and_gen_vllm.py -------------------------------------------------------------------------------- /inference/vllm/monet_gpu_model_runner.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/inference/vllm/monet_gpu_model_runner.py -------------------------------------------------------------------------------- /inference/vllm_inference_example.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/inference/vllm_inference_example.py -------------------------------------------------------------------------------- /monet_qwen_model/apply_qwen2_5_monet.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/monet_qwen_model/apply_qwen2_5_monet.py -------------------------------------------------------------------------------- /monet_qwen_model/modeling_qwen2_5_vl_monet.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/monet_qwen_model/modeling_qwen2_5_vl_monet.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/requirements.txt -------------------------------------------------------------------------------- /script_examples/sft_stage1.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/script_examples/sft_stage1.sh -------------------------------------------------------------------------------- /script_examples/sft_stage2.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/script_examples/sft_stage2.sh -------------------------------------------------------------------------------- /script_examples/sft_stage3.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/script_examples/sft_stage3.sh -------------------------------------------------------------------------------- /src/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/src/main.py -------------------------------------------------------------------------------- /src/precompute_teacher_latents.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/src/precompute_teacher_latents.py -------------------------------------------------------------------------------- /src/precompute_teacher_reps.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/src/precompute_teacher_reps.py -------------------------------------------------------------------------------- /src/task.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/src/task.py -------------------------------------------------------------------------------- /src/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/src/trainer.py -------------------------------------------------------------------------------- /src/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/src/utils.py -------------------------------------------------------------------------------- /transformers/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NOVAglow646/Monet/HEAD/transformers/.DS_Store --------------------------------------------------------------------------------