Papers › Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones?

Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones?

18 Jun 2024arXiv:2406.12809archive 2025-07-28

Zhe Yang, Yichang Zhang, Tianyu Liu, Jian Yang, Junyang Lin, Chang Zhou, Zhifang Sui

Large language models (LLMs) have demonstrated impressive capabilities, but still suffer from inconsistency issues (e.g. LLMs can react differently to disturbances like rephrasing or inconsequential order change). In addition to these inconsistencies, we also observe that LLMs, while capable of solving hard problems, can paradoxically fail at easier ones. To evaluate this hard-to-easy inconsistency, we develop the ConsisEval benchmark, where each entry comprises a pair of questions with a strict order of difficulty. Furthermore, we introduce the concept of consistency score to quantitatively measure this inconsistency and analyze the potential for improvement in consistency by relative consistency score. Based on comprehensive experiments across a variety of existing models, we find: (1) GPT-4 achieves the highest consistency score of 92.2% but is still inconsistent to specific questions due to distraction by redundant information, misinterpretation of questions, etc.; (2) models with stronger capabilities typically exhibit higher consistency, but exceptions also exist; (3) hard data enhances consistency for both fine-tuning and in-context learning. Our data and code will be publicly available on GitHub.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 14 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 2 honoured, 1 violated, 4 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 14 samples from 1 repository, 11 ran (7 with no instrument failure; 4 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

QwenLM/ConsisEval officialmentioned on GitHubMITSyntology: 14 samples, 11 ran (4 instrument-failed) · 3 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

14 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 2 honoured, 1 violated, 4 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
1ran · violated contract
4ran · our draft was wrong
4ran
3unverified

Licence: 4 of the 14 samples are pointer only, meaning Syntology does not serve that copy's text. For 4 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from QwenLM/ConsisEval. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

QwenLM/ConsisEval official repository 14 samples, 11 ran (4 instrument-failed) · 3 unverified
are_equal_under_sympy in 7 papers math_check/grader.py ran fingerprinted MIT (permissive) · 8bec43a95385f997 · report fixture: signature ground_truth_normalized:str/0/str, given_normalized:str/0/str · returned bool · level recorded 2026-09-24
check_correctness_code code_check/check.py ran MIT (permissive) · b0505332a16b5516 · report fixture: signature problem:dict/0/?, completion:str/0/str, timeout:float/0/float, completion_id:NoneType/0/? · returned bool · level recorded 2026-09-24
conflict_make in 7 papers instruction_following_check/instructions_registry.py ran · our draft was wrong MIT (permissive) · d197da35ee544f75 · report fixture: signature conflicts:dict/0/? · returned dict · level recorded 2026-09-24
count_unknown_letters_in_expr in 14 papers math_check/grader.py ran · honoured contract fingerprinted MIT (permissive) · 54417bf5eb16e81f · report fixture: signature expr:str/0/str · returned int · level recorded 2026-09-24
count_words in 28 papers instruction_following_check/instructions_util.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · cdcc85ca09b00f7e · report fixture: signature text:str/0/str · returned int · level recorded 2026-09-24
extract_answer in 112 papers math_check/check.py ran fingerprinted MIT (permissive) · b22b268ce5bb5b68 · report fixture: signature last_sentence:str/0/str, prompt:str/0/str · returned tuple · level recorded 2026-09-24
normalize_answer in 304 papers math_check/math_normalize.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · 101b1e54fcdadf90 · report fixture: signature answer:str/0/str · returned str · level recorded 2026-09-24
process_text in 21 papers code_check/post_process.py ran fingerprinted MIT (permissive) · dab90bb25f71a531 · report fixture: signature text:str/0/str, entry_point:str/0/str · returned str · level recorded 2026-09-24
remove_after_return in 5 papers code_check/post_process.py ran · our draft was wrong fingerprinted MIT (permissive) · 29ee1d8c120f674a · report fixture: signature code:str/0/str · returned str · level recorded 2026-09-24
should_allow_eval in 14 papers math_check/grader.py ran · violated contract MIT (permissive) · f5c7c44887cd8c88 · report fixture: signature expr:str/0/str · returned bool · level recorded 2026-09-24
split_into_sentences in 18 papers instruction_following_check/instructions_util.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · 34d346853108011e · report fixture: signature text:str/0/str · returned list · level recorded 2026-09-24
check_correctness_instruction_following instruction_following_check/check.py unverified MIT (permissive) · f9e0e08dca784150 · report
check_correctness_math math_check/check.py unverified MIT (permissive) · a13ec071b5563491 · report
count_sentences in 13 papers instruction_following_check/instructions_util.py unverified MIT (permissive) here; the stored copy is pointer only · f79646ddd0a03162 · report

Tasks

In-Context Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGPT-4Label SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections