SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.14016 · EMNLP · 2024

Is LLM-as-a-Judge Robust? Investigating Universal Adversarial Attacks on Zero-shot LLM Assessment

Vyas Raina, Mark Gales, Adian Liusie

arXiv · PDF · Open in the Atlas

Code that ran

3 of 7 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, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 4 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/2402.14016.json · badge: /api/ran/2402.14016.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
rainavyas/attack-comparative-assessment marked official in Papers with Code
Papers with Code archive
3 ran · 4 no recorded run
FunctionStatusWhere it lives
comparative_evals Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rainavyas/attack-comparative-assessment/src/attacker/greedy.py
pointer only (licence: NONE) · get_code("c751c48283f89ab3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, rainavyas/attack-comparative-assessment 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.

load_prompt_template Ran
ran; our contract draft was wrong, not the code
rainavyas/attack-comparative-assessment/src/attacker/greedy.py
pointer only (licence: NONE) · get_code("42ad8d850594ddc0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, rainavyas/attack-comparative-assessment 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.

next_dir Ran
ran; our contract draft was wrong, not the code
rainavyas/attack-comparative-assessment/src/attacker/greedy.py
pointer only (licence: NONE) · get_code("7ef950ed1649e351")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, rainavyas/attack-comparative-assessment 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.

BaseAttacker Not yet run
unverified (harvested, no recorded run)
rainavyas/attack-comparative-assessment/src/attacker/greedy.py
pointer only (licence: NONE) · get_code("01fed6edf77646d6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, rainavyas/attack-comparative-assessment 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.

BaseComparativeAttacker Not yet run
unverified (harvested, no recorded run)
rainavyas/attack-comparative-assessment/src/attacker/greedy.py
pointer only (licence: NONE) · get_code("393ab376c804510c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, rainavyas/attack-comparative-assessment 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.

BaseGreedyAttacker Not yet run
unverified (harvested, no recorded run)
rainavyas/attack-comparative-assessment/src/attacker/greedy.py
pointer only (licence: NONE) · get_code("a77ff2accb07cdfb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, rainavyas/attack-comparative-assessment 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.

GreedyComparativeAttacker Not yet run
unverified (harvested, no recorded run)
rainavyas/attack-comparative-assessment/src/attacker/greedy.py
pointer only (licence: NONE) · get_code("aefa45d1e5d9d27e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, rainavyas/attack-comparative-assessment 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.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 7 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.

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

Large Language Models (LLMs) are powerful zero-shot assessors used in real-world situations such as assessing written exams and benchmarking systems. Despite these critical applications, no existing work has analyzed the vulnerability of judge-LLMs to adversarial manipulation. This work presents the first study on the adversarial robustness of assessment LLMs, where we demonstrate that short universal adversarial phrases can be concatenated to deceive judge LLMs to predict inflated scores. Since adversaries may not know or have access to the judge-LLMs, we propose a simple surrogate attack where a surrogate model is first attacked, and the learned attack phrase then transferred to unknown judge-LLMs. We propose a practical algorithm to determine the short universal attack phrases and demonstrate that when transferred to unseen models, scores can be drastically inflated such that irrespective of the assessed text, maximum scores are predicted. It is found that judge-LLMs are significantly more susceptible to these adversarial attacks when used for absolute scoring, as opposed to comparative assessment. Our findings raise concerns on the reliability of LLMas-a-judge methods, and emphasize the importance of addressing vulnerabilities in LLM assessment methods before deployment in highstakes real-world scenarios. 1

For agents

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

get_harvested_code_for_paper("2402.14016")
get_code_for_paper("2402.14016")
have("2402.14016")

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

curl https://syntology.ai/api/ran/2402.14016.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.14016.svg)](https://syntology.ai/paper/2402.14016)

Connect an agent — have() is free.