├── .gitignore ├── LICENSE ├── README.md ├── configs ├── mcts_round1.yaml ├── mcts_sft.yaml ├── mcts_sft_round.yaml ├── offline_inference.yaml ├── react_demo.yaml ├── react_round1.yaml ├── react_sft.yaml ├── sbs_greedy.yaml └── sbs_sft.yaml ├── eval_output_jsonl.py ├── imgs ├── Q_distribution.png ├── Q_distribution_test.png ├── mcts.png ├── mcts_example.pdf └── mcts_example.png ├── implementation_details.md ├── mcts_math ├── __init__.py ├── agents │ ├── __init__.py │ ├── mcts.py │ ├── react.py │ ├── step_beam.py │ ├── tree.py │ └── utils.py ├── config.py ├── constants.py ├── few_shots │ ├── gsm8k.json │ ├── math.json │ └── prompt.json ├── llms │ ├── __init__.py │ ├── local_llm_engine.py │ └── local_llms.py ├── nodes │ ├── __init__.py │ ├── base_node.py │ └── mcts_node.py ├── prompts │ ├── __init__.py │ ├── prompt_react.py │ └── prompt_sft.py ├── react_batch.py ├── solver.py └── tools │ ├── __init__.py │ └── python_tool.py ├── offline_inference.py ├── react_batch_demo.py ├── react_demo.py ├── requirements.txt ├── run_greedy.sh ├── run_sbs.sh ├── scripts ├── modeling_value_head.py └── save_value_head.py └── solver_demo.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/.gitignore -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/README.md -------------------------------------------------------------------------------- /configs/mcts_round1.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/mcts_round1.yaml -------------------------------------------------------------------------------- /configs/mcts_sft.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/mcts_sft.yaml -------------------------------------------------------------------------------- /configs/mcts_sft_round.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/mcts_sft_round.yaml -------------------------------------------------------------------------------- /configs/offline_inference.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/offline_inference.yaml -------------------------------------------------------------------------------- /configs/react_demo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/react_demo.yaml -------------------------------------------------------------------------------- /configs/react_round1.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/react_round1.yaml -------------------------------------------------------------------------------- /configs/react_sft.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/react_sft.yaml -------------------------------------------------------------------------------- /configs/sbs_greedy.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/sbs_greedy.yaml -------------------------------------------------------------------------------- /configs/sbs_sft.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/configs/sbs_sft.yaml -------------------------------------------------------------------------------- /eval_output_jsonl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/eval_output_jsonl.py -------------------------------------------------------------------------------- /imgs/Q_distribution.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/imgs/Q_distribution.png -------------------------------------------------------------------------------- /imgs/Q_distribution_test.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/imgs/Q_distribution_test.png -------------------------------------------------------------------------------- /imgs/mcts.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/imgs/mcts.png -------------------------------------------------------------------------------- /imgs/mcts_example.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/imgs/mcts_example.pdf -------------------------------------------------------------------------------- /imgs/mcts_example.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/imgs/mcts_example.png -------------------------------------------------------------------------------- /implementation_details.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/implementation_details.md -------------------------------------------------------------------------------- /mcts_math/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /mcts_math/agents/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/agents/__init__.py -------------------------------------------------------------------------------- /mcts_math/agents/mcts.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/agents/mcts.py -------------------------------------------------------------------------------- /mcts_math/agents/react.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/agents/react.py -------------------------------------------------------------------------------- /mcts_math/agents/step_beam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/agents/step_beam.py -------------------------------------------------------------------------------- /mcts_math/agents/tree.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/agents/tree.py -------------------------------------------------------------------------------- /mcts_math/agents/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/agents/utils.py -------------------------------------------------------------------------------- /mcts_math/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/config.py -------------------------------------------------------------------------------- /mcts_math/constants.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/constants.py -------------------------------------------------------------------------------- /mcts_math/few_shots/gsm8k.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/few_shots/gsm8k.json -------------------------------------------------------------------------------- /mcts_math/few_shots/math.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/few_shots/math.json -------------------------------------------------------------------------------- /mcts_math/few_shots/prompt.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/few_shots/prompt.json -------------------------------------------------------------------------------- /mcts_math/llms/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/llms/__init__.py -------------------------------------------------------------------------------- /mcts_math/llms/local_llm_engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/llms/local_llm_engine.py -------------------------------------------------------------------------------- /mcts_math/llms/local_llms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/llms/local_llms.py -------------------------------------------------------------------------------- /mcts_math/nodes/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/nodes/__init__.py -------------------------------------------------------------------------------- /mcts_math/nodes/base_node.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/nodes/base_node.py -------------------------------------------------------------------------------- /mcts_math/nodes/mcts_node.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/nodes/mcts_node.py -------------------------------------------------------------------------------- /mcts_math/prompts/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /mcts_math/prompts/prompt_react.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/prompts/prompt_react.py -------------------------------------------------------------------------------- /mcts_math/prompts/prompt_sft.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/prompts/prompt_sft.py -------------------------------------------------------------------------------- /mcts_math/react_batch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/react_batch.py -------------------------------------------------------------------------------- /mcts_math/solver.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/solver.py -------------------------------------------------------------------------------- /mcts_math/tools/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/tools/__init__.py -------------------------------------------------------------------------------- /mcts_math/tools/python_tool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/mcts_math/tools/python_tool.py -------------------------------------------------------------------------------- /offline_inference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/offline_inference.py -------------------------------------------------------------------------------- /react_batch_demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/react_batch_demo.py -------------------------------------------------------------------------------- /react_demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/react_demo.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | antlr4-python3-runtime==4.11.1 2 | omegaconf==2.4.0.dev2 3 | pebble==5.0.7 4 | 5 | -------------------------------------------------------------------------------- /run_greedy.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/run_greedy.sh -------------------------------------------------------------------------------- /run_sbs.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/run_sbs.sh -------------------------------------------------------------------------------- /scripts/modeling_value_head.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/scripts/modeling_value_head.py -------------------------------------------------------------------------------- /scripts/save_value_head.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/scripts/save_value_head.py -------------------------------------------------------------------------------- /solver_demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MARIO-Math-Reasoning/Super_MARIO/HEAD/solver_demo.py --------------------------------------------------------------------------------