SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2607.28478 · 2026

Would You Walk to the Car Wash? Revealing the Salience Bias of Large Language Models in Commonsense Reasoning

Cheng Yang, Chenhao Xue, Zhuosheng Zhang, Zheng Wu, Shijie Zheng, Yijie Lu

arXiv · PDF · Open in the Atlas

Code that ran

13 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, 0 violated it and 0 have no contract verdict; 13 failures of our instrument; 2 no recorded run), measured 2026-11-22; 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/2607.28478.json · badge: /api/ran/2607.28478.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
Wuzheng02/SaliTrap canonical (classifier label)
classifier basis: census
13 ran · 2 no recorded run
FunctionStatusWhere it lives
build_output_record Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/run_pipeline.py
pointer only (licence: NONE) · get_code("9caa956054b1f540")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

call_judge Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Wuzheng02/SaliTrap/eval/eval_phystrap.py
pointer only (licence: NONE) · get_code("6d59380bef1ea165")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

json_block Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/stability_retest/stability_retest.py
pointer only (licence: NONE) · get_code("ddb78bd228e533c2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

json_block Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/v3_registry/phystrap_v3_candidate_registry.py
pointer only (licence: NONE) · get_code("041d3b8ebc3f5027")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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_checkpoint Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/run_pipeline.py
pointer only (licence: NONE) · get_code("6d898cca135ec738")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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_module Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/run_pipeline.py
pointer only (licence: NONE) · get_code("0930c9d5476d5050")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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_module Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/v3_registry/phystrap_v3_candidate_registry.py
pointer only (licence: NONE) · get_code("f8906b04cc90f9e7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

md_code Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/stability_retest/stability_retest.py
pointer only (licence: NONE) · get_code("229c159a6e0bf973")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

normalize_label Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/stability_retest/stability_retest.py
pointer only (licence: NONE) · get_code("7ea76b00e4663067")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

normalize_label Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/v2_pipeline/phystrap_filtering_pipeline_v2.py
pointer only (licence: NONE) · get_code("39dc182862e95f62")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

normalize_seed Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/seed_scaler/phystrap_seed_scaler.py
pointer only (licence: NONE) · get_code("3a35459ca09e78b4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

parse_json_array_from_text Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/seed_scaler/phystrap_seed_scaler.py
pointer only (licence: NONE) · get_code("d7d26bd025b12a4d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

validate_seed_structure Ran
ran; our contract draft was wrong, not the code
Wuzheng02/SaliTrap/seed_scaler/phystrap_seed_scaler.py
pointer only (licence: NONE) · get_code("3000b220e3c227fc")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

call_solver Not yet run
unverified (harvested, no recorded run)
Wuzheng02/SaliTrap/eval/eval_phystrap.py
pointer only (licence: NONE) · get_code("dcc19513417e1fc7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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.

call_with_retry Not yet run
unverified (harvested, no recorded run)
Wuzheng02/SaliTrap/eval/eval_phystrap.py
pointer only (licence: NONE) · get_code("42cd2412fa4ca01c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Wuzheng02/SaliTrap 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 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, 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 15 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

As large language models (LLMs) continue to advance in complex reasoning tasks, they have learned to heavily prioritize explicit conditions provided in the input. However, in everyday commonsense reasoning, this mechanism exposes a critical vulnerability which we term Salience Bias: models become easily hijacked by useless explicit distractors (e.g., numerical values), leading them to ignore the implicit physical or commonsense prerequisites of a task. A critical open question is whether this failure reflects a genuine gap in commonsense knowledge or merely its suppression under misleading task framing. To investigate this, we construct the SaliTrap Benchmark, a high-quality dataset across four trap dimensions. Evaluating 12 state-of-the-art LLMs, we find that all mainstream models suffer significantly from salience bias, with severity scaling with distractor density and detecting the trap often decoupled from actually avoiding it. Crucially, by re-eliciting the same models with the task framing stripped away, we show that this is overwhelmingly a failure of knowledge suppression rather than knowledge absence: a context-free knowledge probe alone recovers over 90% of sycophantic-compliance failures, revealing that the requisite commonsense is intrinsically present but actively crowded out by salient distractors that lure the model into over-compliant, unnecessary computation. Building on this diagnosis, we further show that lightweight, inference-time prompting alone substantially closes the gap without any retraining. Our findings relocate the bottleneck of commonsense reasoning failures from model competence to elicitation, and we release SaliTrap as a testbed for this blind spot. The codes are available at https://github.com/Wuzheng02/SaliTrap.

For agents

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

get_harvested_code_for_paper("2607.28478")
get_code_for_paper("2607.28478")
have("2607.28478")

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

curl https://syntology.ai/api/ran/2607.28478.json

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

[![Syntology run record](https://syntology.ai/api/ran/2607.28478.svg)](https://syntology.ai/paper/2607.28478)

Connect an agent — have() is free.