SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.08326 · EMNLP · 2024

RoTBench: A Multi-Level Benchmark for Evaluating the Robustness of Large Language Models in Tool Learning

Qi Zhang, Xuanjing Huang, Junjie Ye, Tao Gui, Xiaoran Fan, Songyang Gao, Guanyu Li, Yilong Wu, Sixian Li

arXiv · PDF · Open in the Atlas

Code that ran

10 of 15 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 9 failures of our instrument; 5 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/2401.08326.json · badge: /api/ran/2401.08326.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
Junjie-Ye/RoTBench canonical (classifier label)
classifier basis: census
10 ran · 5 no recorded run
FunctionStatusWhere it lives
compress_prompt Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Inference/models.py
code served (permissive licence) · get_code("b8f6a0e07e142f30")
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 re

class StandInTokenizer:
    def encode(self, text):
        # Simple tokenizer: split by spaces and return a list of tokens
        # Each token represented as an integer
        return [i for i in range(len(text.split()))]

# Test for 'completion' mode
prompt_completion = "Function: def add(a, b): return a + b\n\nAssistant: This function adds two numbers." + " extra content" * 2000

tokenizer = StandInTokenizer()

# Test for 'chat' mode
prompt_chat = [
    {"role": "user", "content": "What is this function?"},
    {"role": "function", "content": "Function: def multiply(a, b): return a * b" + " extra" * 1500},
    {"role": "assistant", "content": "This multiplies two numbers."},
    {"role": "function", "content": "Another function: def divide(a, b): return a / b" + " extra" * 1500},
    {"role": "user", "content": "How does it work?"},
]

ARGS = {
    'prompt': prompt_completion,
    'tokenizer': tokenizer,
    'mode': 'completion'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 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
conversations_process Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Data_Process/transform_process.py
code served (permissive licence) · get_code("c6316fc799a85487")
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 list, 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 torch

torch.manual_seed(0)

# Build conversations list with different 'from' types
conversations = [
    {
        'from': 'system',
        'value': 'old system message'
    },
    {
        'from': 'assistant',
        'value': [
            'Action: search_tool\nAction Input: {"query": "test"}',
            'Action: api_call\nAction Input: {"param1": "value1"}'
        ]
    },
    {
        'from': 'function',
        'value': 'Result from search_tool: success'
    },
    {
        'from': 'user',
        'value': 'Use search_tool\nto find results'
    },
    {
        'from': 'assistant',
        'value': [
            'Action: ask_to_user\nAction Input: {}',
            'Action: finish\nAction Input: {}'
        ]
    },
    {
        'from': 'function',
        'value': "Function 'api_call' executed with 'old_param': 'result'"
    }
]

# Build match_dict mapping tool names
match_dict = {
    'search_tool': 'new_search',
    'api_call': {
        'old_param': 'new_param',
        'required': 'mandatory_field',
        'yes': 'default_value'
    }
}

# Build trans_dict for parameter translations
trans_dict = {
    'search_tool': {
        'old_index': 'new_index'
    },
    'api_call': {
        'old_index': 'new_index'
    }
}

# System document replacement
sys_doc = 'This is the new system document'

# Target can be either 'tool' or 'param'
target = 'tool'

ARGS = {
    'conversations': conversations,
    'match_dict': match_dict,
    'trans_dict': trans_dict,
    'sys_doc': sys_doc,
    'target': target
}
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 list, ndim 1, dtype family float — a list 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
get_answer_list Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Evaluation/evaluate.py
code served (permissive licence) · get_code("f6309633583a2931")
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

torch.manual_seed(0)

data = {
    'conversations': [
        {
            'value': 'First message'
        },
        {
            'value': 'Second message'
        },
        {
            'value': 'Final answer'
        }
    ]
}

ARGS = {
    'data': data
}
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
get_cata_list Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Evaluation/evaluate.py
code served (permissive licence) · get_code("31049a1bb3c726a8")
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 list. 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 json
import tempfile
import os

# Create a temporary JSON file with test data
temp_dir = tempfile.mkdtemp()
answer_file = os.path.join(temp_dir, "answers.json")

# Create test data with various scenario types
test_data = [
    {"scenario": "TG", "question": "Generate text", "answer": "sample"},
    {"scenario": "RS", "question": "Search real-time", "answer": "result"},
    {"scenario": "DU", "question": "Understand data", "answer": "insight"},
    {"scenario": "PL", "question": "Personal question", "answer": "response"},
    {"scenario": "AM", "question": "Manipulate app", "answer": "action"},
    {"scenario": "IR", "question": "Retrieve info", "answer": "data"},
    {"scenario": "FT", "question": "Financial query", "answer": "transaction"},
    {"scenario": "TG", "question": "Another text gen", "answer": "output"},
    {"scenario": "RS", "question": "Another search", "answer": "found"},
]

# Write test data to the temporary file
with open(answer_file, 'w', encoding='utf-8') as f:
    json.dump(test_data, f)

ARGS = {
    "answer_file": answer_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 list — a list 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
get_compressed_list Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/model/compression.py
code served (permissive licence) · get_code("709a780a94ca2670")
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 list, 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple module with nested Linear layers
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
        self.nested = nn.Sequential(
            nn.Linear(2, 5),
            nn.ReLU(),
            nn.Linear(5, 1)
        )
    
    def forward(self, x):
        return x

module = SimpleModel()

ARGS = {
    'module': module,
    'prefix': ''
}
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 list, ndim 1, dtype family float — a list 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
mode_control Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Data_Process/transform_process.py
code served (permissive licence) · get_code("b7e2a5655a7335a3")
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, 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 random

def mode_control(mode):
    if 'tool' in mode:
        target_control = 'tool'
    else:
        target_control = 'param'
    tool_control, param_control = (0, 0)
    if mode == 'tool_slight':
        tool_control = 1
    elif mode == 'tool_medium':
        tool_control = 2
    elif mode == 'tool_heavy':
        tool_control = 3
    elif mode == 'uni_tool':
        tool_control = random.choice([1, 2, 3])
    elif mode == 'param_slight':
        param_control = 1
    elif mode == 'param_medium':
        param_control = 2
    elif mode == 'param_heavy':
        param_control = 3
    elif mode == 'uni_param':
        param_control = random.choice([1, 2, 3])
    return (tool_control, param_control, target_control)

ARGS = {
    'mode': 'tool_slight'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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
raise_warning_for_incompatible_cpu_offloading_configuration Ran
ran, violated the contract we drafted
Junjie-Ye/RoTBench/Code/model/model_adapter.py
code served (permissive licence) · get_code("0bf414a2b44babe6")
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 bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape 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 sys
import warnings

# Set up arguments for the function
ARGS = {
    "device": "cuda",
    "load_8bit": True,
    "cpu_offloading": True,
}
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 bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
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
random_name Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Data_Process/transformer.py
code served (permissive licence) · get_code("b5daeb59f8c9cfd0")
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 random
import string

# Set seed for reproducibility
random.seed(0)

# The function random_name takes two parameters: min and max
# It uses them to create a range and select a random count within that range
# Then generates a random string of that length

ARGS = {
    'min': 3,
    'max': 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 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
reverse Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Data_Process/transformer.py
code served (permissive licence) · get_code("78e64427cacb1728")
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

def reverse(str):
    return str[::-1]

ARGS = {
    "str": "hello"
}
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
tool_slight_transform Ran
ran; our contract draft was wrong, not the code
Junjie-Ye/RoTBench/Code/Data_Process/transformer.py
code served (permissive licence) · get_code("9f30e5c30d04940e")
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 random
import string
from math import ceil

# Set seed for reproducibility
random.seed(42)

# Create a minimal config that the function expects
# The function iterates over config and accesses tool['name']
config = [
    {'name': 'hammer'},
    {'name': 'screwdriver'},
    {'name': 'wrench'},
    {'name': 'pliers'},
]

# Create a function parameter - just a string that will be in the tool_change_list
function = 'hammer'

ARGS = {
    'config': config,
    'function': function,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 — 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
compress Not yet run
unverified (harvested, no recorded run)
Junjie-Ye/RoTBench/Code/model/compression.py
code served (permissive licence) · get_code("386536152177a379")
How it ran

no invocation recorded.

get_config Not yet run
unverified (harvested, no recorded run)
Junjie-Ye/RoTBench/Code/Evaluation/evaluate.py
code served (permissive licence) · get_code("bbfc0fbb6c40d003")
How it ran

no invocation recorded.

get_conv_template Not yet run
unverified (harvested, no recorded run)
Junjie-Ye/RoTBench/Code/tool_conversation.py
code served (permissive licence) · get_code("6aa619d359276223")
How it ran

no invocation recorded.

get_gpu_memory Not yet run
unverified (harvested, no recorded run)
Junjie-Ye/RoTBench/Code/utils.py
code served (permissive licence) · get_code("2f3cb5d0dcc18502")
How it ran

no invocation recorded.

load_compress_model Not yet run
unverified (harvested, no recorded run)
Junjie-Ye/RoTBench/Code/model/compression.py
code served (permissive licence) · get_code("5f9d4d17fb70af2b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 1 the same month, 6 later) and 9 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
get_conv_template 69 other papers
14 dated earlier, 1 the same month, 49 later, 4 same year, order not determinable, 1 undated
1911.03705 (2019-11)
CommonGen: A Constrained Text Generation Challenge for Generative Commonsense Reasoning
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2204.07705 (2022-04)
SUPER-NATURALINSTRUCTIONS: Generalization via Declarative Instructions on 1600+ NLP Tasks
and 66 more not named here
raise_warning_for_incompatible_cpu_offloading_configuration 15 other papers
4 dated earlier, 11 later
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2310.13023 (2023-10)
GraphGPT: Graph Instruction Tuning for Large Language Models
2312.00401 (2023-12)
VIoTGPT: Learning to Schedule Vision Tools in LLMs towards Intelligent Video Internet of Things
and 12 more not named here
get_gpu_memory 7 other papers
6 dated earlier, 1 later
2305.10973 (2023-05)
Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold
2307.12981 (2023-07)
3D-LLM: Injecting the 3D World into Large Language Models
2308.03279 (2023-08)
UniversalNER: TARGETED DISTILLATION FROM LARGE LANGUAGE MODELS FOR OPEN NAMED ENTITY RECOGNITION
and 4 more not named here
compress 2 other papers
1 dated earlier, 1 later
2312.00401 (2023-12)
VIoTGPT: Learning to Schedule Vision Tools in LLMs towards Intelligent Video Internet of Things
2403.07714 (2024-03)
StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models
get_compressed_list 2 other papers
1 dated earlier, 1 later
2312.00401 (2023-12)
VIoTGPT: Learning to Schedule Vision Tools in LLMs towards Intelligent Video Internet of Things
2403.07714 (2024-03)
StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models
load_compress_model 2 other papers
1 dated earlier, 1 later
2312.00401 (2023-12)
VIoTGPT: Learning to Schedule Vision Tools in LLMs towards Intelligent Video Internet of Things
2403.07714 (2024-03)
StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models

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

Tool learning has generated widespread interest as a vital means of interaction between Large Language Models (LLMs) and the physical world. Current research predominantly emphasizes LLMs' capacity to utilize tools in well-structured environments while overlooking their stability when confronted with the inevitable noise of the real world. To bridge this gap, we introduce RoTBench, a multi-level benchmark for evaluating the robustness of LLMs in tool learning. Specifically, we establish five external environments, each featuring varying levels of noise (i.e., Clean, Slight, Medium, Heavy, and Union), providing an in-depth analysis of the model's resilience across three critical phases: tool selection, parameter identification, and content filling. Experiments involving six widely-used models underscore the urgent necessity for enhancing the robustness of LLMs in tool learning. For instance, the performance of GPT-4 even drops significantly from 80.00 to 58.10 when there is no substantial change in manual accuracy. More surprisingly, the noise correction capability inherent in the GPT family paradoxically impedes its adaptability in the face of mild noise. In light of these findings, we propose RoTTuning, a strategy that enriches the diversity of training environments to bolster the robustness of LLMs in tool learning. The code and data are available at https: //github.com/Junjie-Ye/RoTBench.

For agents

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

get_harvested_code_for_paper("2401.08326")
get_code_for_paper("2401.08326")
have("2401.08326")

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

curl https://syntology.ai/api/ran/2401.08326.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.08326.svg)](https://syntology.ai/paper/2401.08326)

Connect an agent — have() is free.