SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.03629 · ICLR · 2023

ReAct: Synergizing Reasoning and Acting in Language Models

Karthik Narasimhan, Yuan Cao, Nan Du, Shunyu Yao, Jeffrey Zhao, Dian Yu, Izhak Shafran

arXiv · PDF · Open in the Atlas

Code that ran

15 of 34 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (9 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 5 failures of our instrument; 19 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2210.03629.json · badge: /api/ran/2210.03629.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
ysymyth/ReAct canonical (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
sierra-research/tau-bench community
Papers with Code archive
9 ran · 17 no recorded run
liyuan24/nanodeepresearch community
Papers with Code archive
3 ran · 2 no recorded run
FunctionStatusWhere it lives
Action Ran
ran on a synthesized input · constructed an object; no result was computed
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("69febc56d55428d1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, Action: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from typing import Optional, List, Dict, Any, Tuple
from pydantic import BaseModel

class Action(BaseModel):
    name: str
    kwargs: Dict[str, Any]

# Construct valid arguments for Action
ARGS = {
    "name": "forward_pass",
    "kwargs": {
        "batch_size": 32,
        "learning_rate": 0.001,
        "hidden_dim": 128,
        "tensor_data": torch.randn(2, 4),
        "labels": torch.randint(0, 2, (2,)),
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pydantic, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Action — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Action output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
AgentState Ran
ran on a synthesized input · constructed an object; no result was computed
liyuan24/nanodeepresearch/agent/react_agent.py
pointer only (licence: NONE) · get_code("67368f282ad305f4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, liyuan24/nanodeepresearch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

HumanUserSimulationEnv Ran
ran on a synthesized input · constructed an object; no result was computed
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("c06bc83fe26205ee")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, HumanUserSimulationEnv: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from abc import ABC, abstractmethod

class BaseUserSimulationEnv(ABC):
    metadata = {}

    @abstractmethod
    def reset(self, instruction=None):
        raise NotImplementedError

    @abstractmethod
    def step(self, content):
        raise NotImplementedError

    @abstractmethod
    def get_total_cost(self):
        raise NotImplementedError

class HumanUserSimulationEnv(BaseUserSimulationEnv):

    def reset(self, instruction=None):
        return input(f'{instruction}\n')

    def step(self, content):
        return input(f'{content}\n')

    def get_total_cost(self):
        return 0

ARGS = {}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type HumanUserSimulationEnv — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a HumanUserSimulationEnv output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
RewardActionInfo Ran
ran on a synthesized input · constructed an object; no result was computed
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("57d96bea80ca4d0f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, RewardActionInfo: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import hashlib

ARGS = {
    "r_actions": 0.75,
    "gt_data_hash": hashlib.md5(b"test_data").hexdigest(),
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pydantic, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RewardActionInfo — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RewardActionInfo output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
RewardOutputInfo Ran
ran on a synthesized input · constructed an object; no result was computed
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("3cbe32095b32f8a4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, RewardOutputInfo: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from typing import Optional, List, Dict, Any, Tuple
from pydantic import BaseModel

class RewardOutputInfo(BaseModel):
    r_outputs: float
    outputs: Dict[str, bool]

torch.manual_seed(0)

ARGS = {
    "r_outputs": 0.75,
    "outputs": {
        "is_valid": True,
        "is_safe": False,
        "is_helpful": True
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pydantic, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RewardOutputInfo — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RewardOutputInfo output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
SolveResult Ran
ran on a synthesized input · constructed an object; no result was computed
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("585b61f32e449f1f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, SolveResult: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from typing import Optional, List, Dict, Any

torch.manual_seed(0)

ARGS = {
    "reward": 42.5,
    "messages": [
        {"role": "user", "content": "test message 1"},
        {"role": "assistant", "content": "test response 1"}
    ],
    "info": {
        "steps": 10,
        "model_name": "test_model",
        "status": "success"
    },
    "total_cost": 3.14
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pydantic, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SolveResult — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a SolveResult output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
State Ran
ran on a synthesized input · constructed an object; no result was computed
liyuan24/nanodeepresearch/agent/react_agent.py
pointer only (licence: NONE) · get_code("b31daddc6c61beaf")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, liyuan24/nanodeepresearch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Tool Ran
ran on a synthesized input · constructed an object; no result was computed
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("ea6a9a1ba41d476e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, Tool: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from typing import Optional, List, Dict, Any, Tuple
import abc

torch.manual_seed(0)

class Tool(abc.ABC):

    @staticmethod
    def invoke(*args, **kwargs):
        raise NotImplementedError

    @staticmethod
    def get_info() -> dict[str, Any]:
        raise NotImplementedError

class ConcreteTool(Tool):
    
    @staticmethod
    def invoke(*args, **kwargs):
        return {"result": "invoked", "args": args, "kwargs": kwargs}
    
    @staticmethod
    def get_info() -> dict[str, Any]:
        return {
            "name": "ConcreteTool",
            "description": "A concrete implementation of Tool",
            "version": "1.0"
        }

ARGS = {}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tool — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Tool output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Tool Ran
ran on a synthesized input · constructed an object; no result was computed
liyuan24/nanodeepresearch/agent/react_agent.py
pointer only (licence: NONE) · get_code("c1d21e714eedd29c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, liyuan24/nanodeepresearch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

clean_str Ran
ran; our contract draft was wrong, not the code
ysymyth/ReAct/wikienv.py
code served (permissive licence) · get_code("961911bec23c09ff")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

def clean_str(p):
    return p.encode().decode('unicode-escape').encode('latin1').decode('utf-8')

# The function takes a string parameter 'p' and performs encoding/decoding operations
# We need to provide a valid string that can be encoded and decoded through these transformations

# Test with a simple ASCII string
test_string = "hello world"

ARGS = {
    "p": test_string
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
consistent_hash Ran
ran; our contract draft was wrong, not the code
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("525653e8db7bd970")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from typing import Union, Tuple

Hashable = Union[str, int, float, Tuple['Hashable'], Tuple[Tuple[str, 'Hashable']]]

# Test with various hashable types
ARGS = {
    "value": "test_string"
}

# Additional test cases to verify the function works with different hashable types
test_cases = [
    42,
    3.14,
    ("nested", "tuple"),
    (("key", "value"),),
    "hello_world",
]

# Verify ARGS is properly defined with a valid hashable value
assert isinstance(ARGS["value"], (str, int, float, tuple)), "value must be a valid Hashable type"
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
f1_score Ran
ran, honoured the contract we drafted
ysymyth/ReAct/wrappers.py
code served (permissive licence) · get_code("32a5f1733d9a8971")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re
import string
from collections import Counter

def normalize_answer(s):

    def remove_articles(text):
        return re.sub('\\b(a|an|the)\\b', ' ', text)

    def white_space_fix(text):
        return ' '.join(text.split())

    def remove_punc(text):
        exclude = set(string.punctuation)
        return ''.join((ch for ch in text if ch not in exclude))

    def lower(text):
        return text.lower()
    return white_space_fix(remove_articles(remove_punc(lower(s))))

def f1_score(prediction, ground_truth):
    normalized_prediction = normalize_answer(prediction)
    normalized_ground_truth = normalize_answer(ground_truth)
    ZERO_METRIC = (0, 0, 0)
    if normalized_prediction in ['yes', 'no', 'noanswer'] and normalized_prediction != normalized_ground_truth:
        return ZERO_METRIC
    if normalized_ground_truth in ['yes', 'no', 'noanswer'] and normalized_prediction != normalized_ground_truth:
        return ZERO_METRIC
    prediction_tokens = normalized_prediction.split()
    ground_truth_tokens = normalized_ground_truth.split()
    common = Counter(prediction_tokens) & Counter(ground_truth_tokens)
    num_same = sum(common.values())
    if num_same == 0:
        return ZERO_METRIC
    precision = 1.0 * num_same / len(prediction_tokens)
    recall = 1.0 * num_same / len(ground_truth_tokens)
    f1 = 2 * precision * recall / (precision + recall)
    return (f1, precision, recall)

ARGS = {
    "prediction": "The quick brown fox jumps over the lazy dog",
    "ground_truth": "A quick brown fox jumps over a lazy dog"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
map_role_label Ran
ran; our contract draft was wrong, not the code
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("6a6bfc18b5ee6f7d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

# The function map_role_label takes a single string parameter 'role'
# and returns a string based on the input value.
# No torch, numpy, scipy, or complex objects are needed here.

ARGS = {
    'role': 'user'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
normalize_answer Ran
ran; our contract draft was wrong, not the code
ysymyth/ReAct/wrappers.py
pointer only (licence: MIT); the stored copy get_code serves, from shmsw25/GraphRetriever, is pointer only (licence: NONE) · get_code("dae7ab386661a4f4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from shmsw25/GraphRetriever, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

to_hashable Ran
ran; our contract draft was wrong, not the code
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("290b5b60a6956be6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from typing import Union, List, Dict, Set, Tuple

# Test cases for to_hashable function
ARGS = {
    "item": {
        "nested_dict": {"a": 1, "b": 2.5},
        "nested_list": [1, 2, "three"],
        "nested_set": {4, 5, 6},
        "simple_value": "test"
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Agent Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("bdaf067f70d6cb83")
How it ran

no invocation recorded.

BaseUserSimulationEnv Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("e1bd2b500ed5d45e")
How it ran

no invocation recorded.

ChatReActAgent Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("e480e41efb54cc05")
How it ran

no invocation recorded.

Env Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("970c4a357bfe3ed9")
How it ran

no invocation recorded.

EnvInfo Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("747fe2940f9b3de2")
How it ran

no invocation recorded.

EnvResetResponse Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("fdcd7a0695b2e849")
How it ran

no invocation recorded.

EnvResponse Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("a73129421afbde3a")
How it ran

no invocation recorded.

LLMUserSimulationEnv Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("95c8661758537c4a")
How it ran

no invocation recorded.

ReactAgent Not yet run
unverified (harvested, no recorded run)
liyuan24/nanodeepresearch/agent/react_agent.py
pointer only (licence: NONE) · get_code("3b3b1888dc5abd89")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, liyuan24/nanodeepresearch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

ReactUserSimulationEnv Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("82d74b8e85fde515")
How it ran

no invocation recorded.

ReflectionUserSimulationEnv Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("00d88099b2fae6a1")
How it ran

no invocation recorded.

RewardResult Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("8452d36949e4cc0d")
How it ran

no invocation recorded.

Task Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("c6504ae4ebf5bcab")
How it ran

no invocation recorded.

UserStrategy Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("32835ca402509b76")
How it ran

no invocation recorded.

VerifyUserSimulationEnv Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("4c54d2533220cb6d")
How it ran

no invocation recorded.

load_prompt Not yet run
unverified (harvested, no recorded run)
liyuan24/nanodeepresearch/agent/react_agent.py
pointer only (licence: NONE) · get_code("cb56e34b15e5e739")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, liyuan24/nanodeepresearch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

load_user Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("6ec1d4d5147ad9d7")
How it ran

no invocation recorded.

reflect Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("b0d8172e1c1ba444")
How it ran

no invocation recorded.

verify Not yet run
unverified (harvested, no recorded run)
sierra-research/tau-bench/tau_bench/agents/chat_react_agent.py
code served (permissive licence) · get_code("e0ead272d4fc4589")
How it ran

no invocation recorded.

Also present in other papers' code

Of 34 samples lifted from this paper's linked repositories, 34 carry a function identity and 0 do not. Of the 34 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 3 later) and 31 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
normalize_answer 133 other papers
37 dated earlier, 1 the same month, 94 later, 1 undated · 1 only as an assembly brick
1508.06615 (2015-08)
Character-Aware Neural Language Models
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1608.07905 (2016-08)
Machine Comprehension Using Match-LSTM and Answer Pointer
and 130 more not named here
f1_score 19 other papers
5 dated earlier, 14 later
1809.09600 (2018-09)
HOTPOTQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
1905.05460 (2019-05)
Cognitive Graph for Multi-Hop Reading Comprehension at Scale
1911.02896 (2019-11)
Contextualized Sparse Representations for Real-Time Open-Domain Question Answering
and 16 more not named here
clean_str 6 other papers
6 later
2023.findings-emnlp.205 (2023 (year only))
AutoPlan: Automatic Planning of Interactive Decision-Making Tasks With Large Language Models
2305.15064 (2023-05)
AutoPlan: Automatic Planning of Interactive Decision-Making Tasks With Large Language Models
2311.05772 (2023-11)
ADaPT: As-Needed Decomposition and Planning with Language Models
and 3 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

While large language models (LLMs) have demonstrated impressive performance across tasks in language understanding and interactive decision making, their abilities for reasoning (e.g. chain-of-thought prompting) and acting (e.g. action plan generation) have primarily been studied as separate topics. In this paper, we explore the use of LLMs to generate both reasoning traces and task-specific actions in an interleaved manner, allowing for greater synergy between the two: reasoning traces help the model induce, track, and update action plans as well as handle exceptions, while actions allow it to interface with and gather additional information from external sources such as knowledge bases or environments. We apply our approach, named ReAct, to a diverse set of language and decision making tasks and demonstrate its effectiveness over state-of-the-art baselines in addition to improved human interpretability and trustworthiness. Concretely, on question answering (HotpotQA) and fact verification (Fever), ReAct overcomes prevalent issues of hallucination and error propagation in chain-of-thought reasoning by interacting with a simple Wikipedia API, and generating human-like task-solving trajectories that are more interpretable than baselines without reasoning traces. Furthermore, on two interactive decision making benchmarks (ALFWorld and WebShop), ReAct outperforms imitation and reinforcement learning methods by an absolute success rate of 34% and 10% respectively, while being prompted with only one or two in-context examples.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2210.03629")
get_code_for_paper("2210.03629")
have("2210.03629")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2210.03629.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2210.03629.svg)](https://syntology.ai/paper/2210.03629)

Connect an agent — have() is free.