Browse State-of-the-Art › Common Sense Reasoning

Common Sense Reasoning

325 papers with code · 24 benchmarks · 56 datasets archive 2025-07-28

Natural Language ProcessingReasoning

Common sense reasoning tasks are intended to require the model to go beyond pattern recognition. Instead, the model should use "common sense" or world knowledge to make inferences.

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

24 leaderboard tables shown for this task, 24 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 24 until expanded.

Syntology 60 newer papers were judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 3 placed): WinoGrande (31 judged, 1 placed), ARC (Challenge) (21 judged, 0 placed), ARC (Easy) (17 judged, 0 placed), CommonsenseQA (16 judged, 2 placed), BIG-bench (Causal Judgment) (1 judged, 0 placed), SWAG (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
WinoGrande (77 rows) ST-MoE-32B 269B (fine-tuned) ST-MoE: Designing Stable and Transferable Sparse Expert Models code Syntology community repositories only · 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 3 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified Compare
ARC (Challenge) (54 rows) GPT-4 (few-shot, k=25) GPT-4 Technical Report code Syntology community repositories only · 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 2 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
ARC (Easy) (47 rows) ST-MoE-32B 269B (fine-tuned) ST-MoE: Designing Stable and Transferable Sparse Expert Models code Syntology community repositories only · 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 3 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified Compare
ReCoRD (45 rows) Turing NLR v5 XXL 5.4B (fine-tuned) Toward Efficient Language Model Pretraining and Downstream... — — Compare
CommonsenseQA (38 rows) GPT-4o (HPT) Hierarchical Prompting Taxonomy: A Universal Evaluation Framework... code — Compare
PARus (22 rows) Human Benchmark RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark code Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
RuCoS (22 rows) Human Benchmark RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark code Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
RWSD (22 rows) Golden Transformer — — — Compare
BIG-bench (Disambiguation QA) (9 rows) PaLM 2 (few-shot, k=3, Direct) PaLM 2 Technical Report code — Compare
BIG-bench (Causal Judgment) (9 rows) PaLM 2 (few-shot, k=3, Direct) PaLM 2 Technical Report code — Compare
BIG-bench (Date Understanding) (9 rows) PaLM 2 (few-shot, k=3, CoT) PaLM 2 Technical Report code — Compare
BIG-bench (Sports Understanding) (8 rows) PaLM 2(few-shot, k=3, CoT) PaLM 2 Technical Report code — Compare
Event2Mind test (7 rows) ConvNet Event2Mind: Commonsense Inference on Events, Intents, and Reactions — — Compare
Russian Event2Mind (7 rows) araneum word2vec (skipgram) + GRU Event2Mind for Russian: Understanding Emotions and Intents in... code — Compare
SWAG (5 rows) DeBERTalarge DeBERTa: Decoding-enhanced BERT with Disentangled Attention code Syntology official: no sample here; runs from other or unrecorded repositories · 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified Compare
BIG-bench (Winowhy) (4 rows) PaLM-540B (few-shot, k=5) PaLM: Scaling Language Modeling with Pathways code Syntology 32 ran (of which 16 constructed an object rather than computing a result; 24 with no instrument failure: 2 honoured, 1 violated, 21 with no contract checked; 8 where Syntology's instrument failed) · 5 unverified Compare
BIG-bench (Known Unknowns) (3 rows) PaLM-540B (few-shot, k=5) PaLM: Scaling Language Modeling with Pathways code Syntology 32 ran (of which 16 constructed an object rather than computing a result; 24 with no instrument failure: 2 honoured, 1 violated, 21 with no contract checked; 8 where Syntology's instrument failed) · 5 unverified Compare
BIG-bench (Logical Sequence) (2 rows) Chinchilla-70B (few-shot, k=5) Training Compute-Optimal Large Language Models code Syntology 8 ran (of which 3 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified Compare
Event2Mind dev (2 rows) ConvNet Event2Mind: Commonsense Inference on Events, Intents, and Reactions — — Compare
CODAH (1 row) BERT Large CODAH: An Adversarially Authored Question-Answer Dataset for Common Sense code — Compare
CrowdSource QA (1 row) BERT Predicting Subjective Features of Questions of QA Websites using BERT code — Compare
Visual Dialog v0.9 (1 row) NMN [kottur2018visual] Visual Coreference Resolution in Visual Dialog using Neural Module Networks code — Compare
Visual Dialog v0.9 (1 row) PDUN Probabilistic framework for solving Visual Dialog — — Compare
WinoGAViL (1 row) ViLT WinoGAViL: Gamified Association Benchmark to Challenge... code Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

56 datasets whose archive record lists this task, ordered by the archive's paper count. 30 shown of 56 until expanded.

Subtasks archive 2025-07-28

17 subtasks in the archive's task tree.

Papers archive 2025-07-28

archive papers tagged: 939 · with a code link: 325 · where Syntology ran a sample: 114 (95 with a run with no instrument failure, 19 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (114 of 939 tagged: 95 with a run with no instrument failure, 19 where every run was a failure of Syntology's instrument)

Page 1 of 10: papers 1 to 100 of 939, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 57 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

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