Browse › Natural Language Processing › Common Sense Reasoning › ARC (Challenge)

Common Sense Reasoning archive 2025-07-28

ARC (Challenge) Benchmark (Common Sense Reasoning)

54 rows 41 with code listed 1 metric Dataset page

Syntology Since the archive (Syntology): none placed; 21 newer papers judged to report on this board so far.

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.

The archive carries no text for this table; the description above is the archive's text for the task Common Sense Reasoning. archive 2025-07-28

Over time archive 2025-07-28

Direction inferred from the metric name, not from the archive: Accuracy (higher is better). Points are placed at the row's paper date, not an evaluation date; 52 of 54 rows carry one.

The chart needs JavaScript; the table below carries every value.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 GPT-4 (few-shot, k=25) 96.4 – Paper Code 2023 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 report
2 PaLM 2 (few-shot, CoT, SC) 95.1 – Paper Code 2023 linked, not harvested report
3 Shivaay (4B, few-shot, k=8) 91.04 – – – not matched report
4 StupidLLM 91.03 – – – not matched report
5 Claude 2 (few-shot, k=5) 91 – Paper – 2023 no code linked report
6 Claude 1.3 (few-shot, k=5) 90 – Paper – 2023 no code linked report
7 PaLM 540B (Self Improvement, Self Consistency) 89.8 – Paper – 2022 no code linked report
8 PaLM 540B (Self Consistency) 88.7 – Paper – 2022 no code linked report
9 PaLM 540B (Self Improvement, CoT Prompting) 88.3 – Paper – 2022 no code linked report
10 PaLM 540B (Self Improvement, Standard-Prompting) 87.2 – Paper – 2022 no code linked report
11 PaLM 540B (Standard-Prompting) 87.1 – Paper – 2022 no code linked report
12 ST-MoE-32B 269B (fine-tuned) 86.5 – Paper Code 2022 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 report
13 Claude Instant 1.1 (few-shot, k=5) 85.7 – Paper – 2023 no code linked report
14 GPT-3.5 (few-shot, k=25) 85.2 – Paper Code 2023 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 report
15 PaLM 540B (CoT Prompting) 85.2 – Paper – 2022 no code linked report
16 LLaMA 3 8B + MoSLoRA (fine-tuned) 81.5 – Paper Code 2024 official (archive's flag): 4 ran · 4 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified report
17 LLaMA-3 8B + MixLoRA 79.9 – Paper Code 2024 official (archive's flag): 6 ran · 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 1 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified report
18 LLaMA-2 13B + MixLoRA 69.9 – Paper Code 2024 official (archive's flag): 6 ran · 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 1 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified report
19 PaLM 2-L (1-shot) 69.2 – Paper Code 2023 linked, not harvested report
20 GAL 120B (zero-shot) 67.9 ✓ Paper Code 2022 official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
21 Camelidae-8×34B 65.2 – Paper Code 2024 linked, not harvested report
22 PaLM 2-M (1-shot) 64.9 – Paper Code 2023 linked, not harvested report
23 FLAN 137B (few-shot, k=13) 63.8 – Paper Code 2021 official: harvested for another paper · 0 ran · 1 unverified report
24 FLAN 137B (zero-shot) 63.1 – Paper Code 2021 official: harvested for another paper · 0 ran · 1 unverified report
25 PaLM 2-S (1-shot) 59.6 – Paper Code 2023 linked, not harvested report
26 LLaMA-2 7B + MixLoRA 58.1 – Paper Code 2024 official (archive's flag): 6 ran · 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 1 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified report
27 LLaMA 33B (zero-shot) 57.8 – Paper Code 2023 official: no sample here; runs from other or unrecorded repositories · 37 ran (of which 9 constructed an object rather than computing a result; 25 with no instrument failure: 3 honoured, 0 violated, 22 with no contract checked; 12 where Syntology's instrument failed) · 21 unverified report
28 ST-MoE-L 4.1B (fine-tuned) 56.9 – Paper Code 2022 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 report
29 LLaMA 65B (zero-shot) 56.0 ✓ Paper Code 2023 official: no sample here; runs from other or unrecorded repositories · 37 ran (of which 9 constructed an object rather than computing a result; 25 with no instrument failure: 3 honoured, 0 violated, 22 with no contract checked; 12 where Syntology's instrument failed) · 21 unverified report
30 Mistral 7B (0-shot) 55.5 – Paper Code 2023 official (archive's flag): 9 ran · 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified report
31 GPT-3 175B (1 shot) 53.2 ✓ Paper Code 2020 community repositories only · 45 ran (of which 0 constructed an object rather than computing a result; 37 with no instrument failure: 3 honoured, 2 violated, 32 with no contract checked; 8 where Syntology's instrument failed) · 20 unverified report
32 LLaMA 13B (zero-shot) 52.7 – Paper Code 2023 official: no sample here; runs from other or unrecorded repositories · 37 ran (of which 9 constructed an object rather than computing a result; 25 with no instrument failure: 3 honoured, 0 violated, 22 with no contract checked; 12 where Syntology's instrument failed) · 21 unverified report
33 GPT-3 (zero-shot) 51.4 – Paper Code 2022 official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
34 GPT-3 175B (0-shot) 51.4 – Paper Code 2020 community repositories only · 45 ran (of which 0 constructed an object rather than computing a result; 37 with no instrument failure: 3 honoured, 2 violated, 32 with no contract checked; 8 where Syntology's instrument failed) · 20 unverified report
35 BLOOM 176B (1-shot) 50.85 – Paper Code 2023 linked, not harvested report
36 GLaM 64B/64E (0 shot) 50.3 ✓ Paper – 2021 no code linked report
37 UL2 20B (chain-of-thought + self-consistency) 49.5 – Paper Code 2022 community repositories only · 15 ran (of which 0 constructed an object rather than computing a result; 15 with no instrument failure: 0 honoured, 0 violated, 15 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified report
38 Bloomberg GPT 50B (1-shot) 48.63 – Paper Code 2023 linked, not harvested report
39 GLaM 64B/64E (1 shot) 48.2 ✓ Paper – 2021 no code linked report
40 LLaMA 7B (zero-shot) 47.6 – Paper Code 2023 official: no sample here; runs from other or unrecorded repositories · 37 ran (of which 9 constructed an object rather than computing a result; 25 with no instrument failure: 3 honoured, 0 violated, 22 with no contract checked; 12 where Syntology's instrument failed) · 21 unverified report
41 GPT-NeoX 20B (1-shot) 45.39 – Paper Code 2023 linked, not harvested report
42 phi-1.5-web 1.3B (zero-shot) 44.9 – Paper Code 2023 linked, not harvested report
43 OPT 66B (one-shot) 44.54 – Paper Code 2023 linked, not harvested report
44 OPT-175B 43.94 – Paper Code 2023 official (archive's flag): 2 ran · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified report
45 UL2 20B (chain-of-thought) 42.9 – Paper Code 2022 community repositories only · 15 ran (of which 0 constructed an object rather than computing a result; 15 with no instrument failure: 0 honoured, 0 violated, 15 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified report
46 SparseGPT (175B, 50% Sparsity) 41.3 – Paper Code 2023 official (archive's flag): 2 ran · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified report
47 SparseGPT (175B, 4:8 Sparsity) 39.85 – Paper Code 2023 official (archive's flag): 2 ran · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified report
48 SparseGPT (175B, 2:4 Sparsity) 38.99 – Paper Code 2023 official (archive's flag): 2 ran · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified report
49 Pythia 12B (5-shot) 36.8 – Paper Code 2023 linked, not harvested report
50 BLOOM (few-shot, k=5) 32.9 – Paper Code 2022 official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
51 Pythia 12B (0-shot) 31.8 – Paper Code 2023 linked, not harvested report
52 OPT (few-shot, k=5) 31.1 – Paper Code 2022 official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
53 UL2 20B (zero-shot) 29.8 – Paper Code 2022 community repositories only · 15 ran (of which 0 constructed an object rather than computing a result; 15 with no instrument failure: 0 honoured, 0 violated, 15 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified report
54 OPT-175B (50% Sparsity) 25.6 – Paper Code 2023 official (archive's flag): 2 ran · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified report

All 54 rows shown. 52 link to a paper page on this site; 5 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran 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”: of the code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and U are unverified (harvested, no recorded run). Of the N, C constructed an object rather than computing a result (a part of N, never taken away from it), K ran with no instrument failure (honoured or violated the contract Syntology drafted, or ran with no contract checked) and I are failures of Syntology's instrument, not of the code (a wrong contract draft, or a contract check that could not run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 29 rows have a graph line, from 11 distinct papers. Rows whose paper has a sample that ran: 27 (27 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument). Papers: 10 with a sample that ran: 10 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument. Whose code ran. When the archive marks a repository official for the row's paper, the cell starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code): “official (archive's flag): N ran” counts the samples from it that ran, instrument failures included; “official: harvested, nothing ran”; when it has no sample for this paper and other samples ran, “community repositories only” if every one came from a repository the archive lists for the paper but neither marks official nor records as mentioned in it, or one Syntology harvested outside the archive's links, and “official: no sample here; runs from other or unrecorded repositories” if any came from a repository named in the paper or found in its text by Syntology, or has no repository recorded; “official: harvested for another paper” (no sample from it is linked to this paper) or “official: not harvested”. No state is printed when the archive marks none official. The cell's tooltip names the repositories the samples that ran came from. Of the 29 rows with a graph line, 29 have an archive-official repository: on 14 a sample from it ran (9 with a run with no instrument failure), on 9 the samples that ran came from community repositories only, and on 4 from other or unrecorded repositories. The filters above hide rows and never renumber them; the first two count, by default, only a run with no instrument failure, and offer any run, instrument failures included. Counting each paper once, of 192 samples: 138 ran (of which 9 constructed an object rather than computing a result; 108 with no instrument failure: 10 honoured, 5 violated, 93 with no contract checked; 30 where Syntology's instrument failed) · 54 unverified. Separately, 34 of those 192 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. Per-sample status is on the paper page; each Ran cell links to that paper's sample list. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Since the archive Syntology

Of the 8,888 newer papers whose tables Syntology has read (8,886 checked in full and 2 with a leaderboard still awaiting a verdict), 37 had this board among the leaderboards the extractor considered for their tables (it keeps at most 10 boards per paper and 4 per dataset; the lists of 1,565 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and 21 proposed a result; of those 21, 15 gave no cell that passed the extractor's own checks, 6 were rejected by the independent check, 0 await a verdict and 0 were placed. Syntology has checked 8,886 of the 18,637 papers on this site that are newer than the archive (for 3,717 of them no archive leaderboard matched the paper's tables, so there was nothing further to check); 733 were read and have nothing to place (arXiv has no HTML version of the paper, or that version has no tables), 41 could not be read (the extractor's reply could not be parsed), and results from the other 8,977 appear after they are checked.

Newer papers whose tables Claude Sonnet 4.5 judged to report on this benchmark (a model's judgement, not a result; newest first by arXiv id)

Paper arXiv date Ran Syntology Outcome
ChainDoRA: Tensor-Train Factorized Weight-Decomposed Low-Rank Adaptation for... arXiv:2609.25058 8 Sep 2026 repository linked, no samples harvested refused by a rule: the table does not name the leaderboard's dataset
ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search arXiv:2609.13356 11 Sep 2026 9 harvested: 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 4 honoured, 0 violated, 0 with no contract checked; 5 where Syntology's instrument failed) · 0 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Target-Aware Calibration Data Selection for Preserving Uncertainty in Quantized Language Models arXiv:2608.21019 21 Aug 2026 8 harvested: 7 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; 7 where Syntology's instrument failed) · 1 unverified refused by a rule: the row label the model quoted is not the row's label
Gated Recurrent Transformers: Expressive Depth through Recurrent Modulation arXiv:2608.15062 15 Aug 2026 6 harvested: 6 ran (of which 6 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; every one of the 6 samples that ran constructed an object rather than computing a result refused by a rule: the number the model quoted is not the number printed in the cell it pointed at
Learning to Select, Not Relearn: Hard-Routed Mixtures of Reasoning LoRAs arXiv:2606.31413 30 Jun 2026 3 harvested: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified refused by a rule: the table's column header does not name the leaderboard's metric
ATMA: Long-Context Language Modeling via Polar Attention and Gated-Delta Compression Memory arXiv:2606.25156 23 Jun 2026 9 harvested: 6 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified rejected by the independent check
Variable-Width Transformers arXiv:2606.18246 16 Jun 2026 2 harvested: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified refused by a rule: the table does not name the leaderboard's dataset
Muonᵖ: Muon with Fractional Spectral Powers arXiv:2606.13867 11 Jun 2026 held, no recorded run rejected by the independent check
Gradient-Guided Reward Optimization for Inference-time Alignment arXiv:2606.09635 8 Jun 2026 held, no recorded run refused by a rule: one configuration gave two values for one column
CART: Context-Anchored Recurrent Transformer -- A Parameter-Efficient Architecture... arXiv:2606.01495 31 May 2026 16 harvested: 7 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified rejected by the independent check
TRACE: Toulmin-based Reasoning Assessment through Constructive Elements for LLM CoT Evaluation arXiv:2605.29656 28 May 2026 4 harvested: 4 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; 1 where Syntology's instrument failed) · 0 unverified rejected by the independent check
Echoes within the Reasoning: Stealthy and Effective Watermarking via Chain of Thought arXiv:2605.28890 27 May 2026 repository linked, no samples harvested refused by a rule: the number the model quoted is not the number printed in the cell it pointed at
WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points arXiv:2605.17471 17 May 2026 4 harvested: 2 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 2 samples that ran constructed an object rather than computing a result refused by a rule: the table does not name the leaderboard's dataset
LaTER: Efficient Test-Time Reasoning via Latent Exploration and Explicit Verification arXiv:2605.07315 8 May 2026 20 harvested: 18 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 1 violated, 2 with no contract checked; 14 where Syntology's instrument failed) · 2 unverified rejected by the independent check
Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model... arXiv:2604.12374 14 Apr 2026 24 harvested: 20 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 3 violated, 4 with no contract checked; 12 where Syntology's instrument failed) · 4 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Optimal Decay Spectra for Linear Recurrences arXiv:2604.07658 8 Apr 2026 5 harvested: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified refused by a rule: the row is an ablation variant, not the method itself
Diet Your LLM: Dimension-wise Global Pruning of LLMs via Merging Task-specific Importance Score arXiv:2603.23985 25 Mar 2026 repository linked, no samples harvested refused by a rule: the table does not name the leaderboard's dataset
Open-Source Reproduction and Explainability Analysis of Corrective Retrieval Augmented... arXiv:2603.16169 17 Mar 2026 repository linked, no samples harvested rejected by the independent check
HESTIA: A Hessian-Guided Differentiable Quantization-Aware Training Framework for... arXiv:2601.20745 28 Jan 2026 6 harvested: 4 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models arXiv:2601.07372 12 Jan 2026 8 harvested: 6 ran (of which 4 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified refused by a rule: the table's column header does not name the leaderboard's metric
AnyBCQ: Hardware Efficient Flexible Binary-Coded Quantization for Multi-Precision LLMs arXiv:2510.10467 12 Oct 2025 27 harvested: 12 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 8 where Syntology's instrument failed) · 15 unverified refused by a rule: one configuration gave two values for one column

Syntology 21 newer papers judged to report on this board: the papers counted above as having proposed a result (0 placed); every one is listed here and in the JSON twin. The order is the arXiv identifier's, newest first, not a rank. The judgement is the model's, read from the paper's own tables; the outcome is what Syntology's checks did with it, and no value from a paper that was not placed is printed. “refused by a rule” names the extractor's rule; “no cell passed the extractor's checks” means the extractor's record holds no cell for this board, passed or refused. The 15 counted above as giving no cell that passed the extractor's own checks are the 15 marked “refused by a rule”. Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

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