Browse › Natural Language Processing › Mathematical Reasoning › AIME24

Mathematical Reasoning archive 2025-07-28

AIME24 Benchmark (Mathematical Reasoning)

9 rows 5 with code listed 1 metric

Syntology Since the archive (Syntology): 2 placed; 42 newer papers judged to report on this board so far.

Over time archive 2025-07-28

Direction inferred from the metric name, not from the archive: Acc (higher is better). Points are placed at the row's paper date, not an evaluation date; 5 of 9 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 Xolver 94.4 – Paper Code 2025 linked, not harvested report
2 DeepSeek-r1 79.8 – Paper Code 2025 linked, not harvested report
3 Openai-o1 74.4 – – – not matched report
4 Openai-o1-mini 70.0 – – – not matched report
5 Search-o1 56.7 – Paper Code 2025 linked, not harvested report
6 s1-32B 56.7 – Paper Code 2025 official (archive's flag): 4 ran · 4 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; 3 where Syntology's instrument failed) · 0 unverified report
7 Openai-o1-preview 44.6 – – – not matched report
8 Qwen2.5-72B-Instruct 23.3 – Paper Code 2024 community repositories only · 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) · 2 unverified report
9 Claude3.5-Sonnet 16 – – – not matched report

All 9 rows shown. 5 link to a paper page on this site; 0 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). 2 rows have a graph line, from 2 distinct papers. Rows whose paper has a sample that ran: 2 (1 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument). Papers: 2 with a sample that ran: 1 with a run with no instrument failure, 1 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 2 rows with a graph line, 2 have an archive-official repository: on 1 a sample from it ran (1 with a run with no instrument failure), on 1 the samples that ran came from community repositories only, and on 0 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 7 samples: 5 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; 4 where Syntology's instrument failed) · 2 unverified. Separately, 1 of those 7 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: results placed by Syntology 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), 44 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 1,587 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and 42 proposed a result; of those 42, 33 gave no cell that passed the extractor's own checks, 7 were rejected by the independent check, 0 await a verdict and 2 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.

Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows above. Measured before publishing, on held-out archive papers: the extractor was given 883 of them and produced output for 881 (the other 2 had no title and were excluded before any model call, so they are not part of this measurement). A model adjudicator (Claude Opus 5.5), blind to which entries were real, judged 108 of 110 accepted entries correct on those held-out archive papers; its own controls: 69 of 70 archive-confirmed entries judged true, 101 of 103 corrupted copies judged false. It was not measured on the newer papers whose results are placed on this site.

Paper Method (configuration) Acc Date Where in the paper Code Syntology Report
Long Chain-of-Thought Compression via Fine-Grained Group Policy Optimization arXiv:2602.10048 Qwen2.5- Math-1.5B FGO 19.3 10 Feb 2026 Table 1, row “Qwen2.5- Math-1.5B FGO”Caption: “Table 1: Comparative results of Vanilla, GRPO and FGO, where bold indicates the…” 6 harvested: 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 2 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified report
TokenSqueeze: Performance-Preserving Compression for Reasoning LLMs arXiv:2511.13223 TokenSqueeze 57.5 17 Nov 2025 Table 3, row “TokenSqueeze”Caption: “Table 3: TokenSqueeze achieves the highest accuracy by reducing both the number…” 20 harvested: 15 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 1 honoured, 3 violated, 5 with no contract checked; 6 where Syntology's instrument failed) · 5 unverified report

Syntology 2 entries, one per paper, newest first by month (the arXiv date, else the month in the arXiv id), then by arXiv id. Each value is the cell text as the paper prints it; under "Where in the paper" is the start of the table's caption as the paper prints it (ranking words shown as […]; a caption carrying TeX this page does not render, or quoting a percentage figure, is left to the hover and says so), and hovering there gives the caption start and each value's column header. Not part of the archive and not in the chart above. Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

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
LastOPD: Taming Collapse in Latent On-Policy Distillation arXiv:2609.28845 23 Sep 2026 repository linked, no samples harvested refused by a rule: one configuration gave two values for one column
Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs arXiv:2608.09542 10 Aug 2026 22 harvested: 20 ran (of which 20 constructed an object rather than computing a result; 20 with no instrument failure: 0 honoured, 0 violated, 20 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 20 samples that ran constructed an object rather than computing a result refused by a rule: the table does not name the leaderboard's dataset
GFlowRL: Scaling Distribution-Matching RL to Large Language Models arXiv:2607.13394 15 Jul 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
Denser ≠ Better: Limits of On-Policy Self-Distillation for Continual Post-Training arXiv:2607.01763 2 Jul 2026 12 harvested: 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 7 unverified refused by a rule: the row label the model quoted is not the row's label
HARD-KV: Head-Adaptive Regularization for Decoding-time KV Compression arXiv:2606.28831 27 Jun 2026 9 harvested: 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 3 honoured, 1 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
GEOALIGN: Geometric Rollout Curation for Robust LLM Reinforcement Learning arXiv:2606.26917 25 Jun 2026 3 harvested: 3 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; 2 where Syntology's instrument failed) · 0 unverified refused by a rule: the table's column header does not name the leaderboard's metric
STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability arXiv:2606.19236 17 Jun 2026 11 harvested: 11 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; 9 where Syntology's instrument failed) · 0 unverified refused by a rule: one configuration gave two values for one column
REVES: REvision and VErification--Augmented Training for Test-Time Scaling arXiv:2606.18910 17 Jun 2026 9 harvested: 6 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; 2 where Syntology's instrument failed) · 3 unverified rejected by the independent check
The Quality-Utility Paradox: Why High-Reward Data Impairs Small Model Mathematical Reasoning arXiv:2606.16152 15 Jun 2026 repository linked, no samples harvested rejected by the independent check
How Does Reasoning Flow? Tracing Attention-Induced Information Flow for Targeted RL in LLMs arXiv:2606.10646 9 Jun 2026 10 harvested: 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified refused by a rule: the row is an ablation variant, not the method itself
DyCon: Dynamic Reasoning Control via Evolving Difficulty Modeling arXiv:2606.07108 5 Jun 2026 1 harvested: 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) · 0 unverified refused by a rule: the table's column header does not name the leaderboard's metric
KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in... arXiv:2606.03458 2 Jun 2026 1 harvested: 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) · 0 unverified refused by a rule: the model gave no value to check against the cell
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
CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic Reasoning arXiv:2605.20075 19 May 2026 11 harvested: 6 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; 6 where Syntology's instrument failed) · 5 unverified rejected by the independent check
Stop When Reasoning Converges: Semantic-Preserving Early Exit for Reasoning Models arXiv:2605.17672 17 May 2026 5 harvested: 5 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 0 unverified rejected by the independent check
Breaking Winner-Takes-All: Cooperative Policy Optimization Improves Diverse LLM Reasoning arXiv:2605.11461 12 May 2026 repository linked, no samples harvested refused by a rule: the table's column header does not name the leaderboard's metric
Hint Tuning: Less Data Makes Better Reasoners arXiv:2605.08665 9 May 2026 15 harvested: 7 ran (of which 4 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 8 unverified refused by a rule: one configuration gave two values for one column
Rubric-based On-policy Distillation arXiv:2605.07396 8 May 2026 9 harvested: 7 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; 6 where Syntology's instrument failed) · 2 unverified refused by a rule: the model gave no value to check against the cell
Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization arXiv:2605.05040 6 May 2026 7 harvested: 7 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 2 violated, 0 with no contract checked; 5 where Syntology's instrument failed) · 0 unverified refused by a rule: the row is an ablation variant, not the method itself
When Less is Enough: Efficient Inference via Collaborative Reasoning arXiv:2605.01111 1 May 2026 12 harvested: 10 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 6 where Syntology's instrument failed) · 2 unverified refused by a rule: the model pointed at a header row
Federation over Text: Insight Sharing for Multi-Agent Reasoning arXiv:2604.16778 18 Apr 2026 20 harvested: 14 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; 13 where Syntology's instrument failed) · 6 unverified refused by a rule: the number the model quoted is not the number printed in the cell it pointed at
Training-Free Test-Time Contrastive Learning for Large Language Models arXiv:2604.13552 15 Apr 2026 14 harvested: 12 ran (of which 5 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 1 violated, 5 with no contract checked; 6 where Syntology's instrument failed) · 2 unverified rejected by the independent check
On the Step Length Confounding in LLM Reasoning Data Selection arXiv:2604.06834 8 Apr 2026 18 harvested: 15 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 1 violated, 2 with no contract checked; 12 where Syntology's instrument failed) · 3 unverified refused by a rule: the row is an ablation variant, not the method itself
ETR: Entropy Trend Reward for Efficient Chain-of-Thought Reasoning arXiv:2604.05355 7 Apr 2026 1 harvested: 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) · 0 unverified refused by a rule: one configuration gave two values for one column
Prune as You Generate: Online Rollout Pruning for Faster and Better RLVR arXiv:2603.24840 25 Mar 2026 repository linked, no samples harvested refused by a rule: the row is an ablation variant, not the method itself
PRISM: A Dual View of LLM Reasoning through Semantic Flow and Latent Computation arXiv:2603.22754 24 Mar 2026 18 harvested: 18 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 4 honoured, 0 violated, 4 with no contract checked; 10 where Syntology's instrument failed) · 0 unverified refused by a rule: the model gave no value to check against the cell
Multi-Agent Debate with Memory Masking arXiv:2603.20215 3 Mar 2026 6 harvested: 5 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; 4 where Syntology's instrument failed) · 1 unverified refused by rules: the quoted value appears more than once in the row; one configuration gave two values for one column
Quasar: Quantized Self-Speculative Acceleration for Rapid Inference via... arXiv:2603.01399 2 Mar 2026 repository linked, no samples harvested refused by a rule: the table's column header does not name the leaderboard's metric
PACE: Prefix-Protected and Difficulty-Aware Compression for Efficient Reasoning arXiv:2602.11639 12 Feb 2026 12 harvested: 12 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 8 where Syntology's instrument failed) · 0 unverified refused by a rule: the number the model quoted is not the number printed in the cell it pointed at
Long Chain-of-Thought Compression via Fine-Grained Group Policy Optimization arXiv:2602.10048 10 Feb 2026 6 harvested: 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 2 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified placed
S3-CoT: Self-Sampled Succinct Reasoning Enables Efficient Chain-of-Thought LLMs arXiv:2602.01982 2 Feb 2026 repository linked, no samples harvested refused by a rule: one configuration gave two values for one column
UCPO: Uncertainty-Aware Policy Optimization arXiv:2601.22648 30 Jan 2026 10 harvested: 6 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 1 violated, 1 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Scaling Embeddings Outperforms Scaling Experts in Language Models arXiv:2601.21204 29 Jan 2026 repository linked, no samples harvested refused by a rule: the table's column header does not name the leaderboard's metric
RelayLLM: Efficient Reasoning via Collaborative Decoding arXiv:2601.05167 8 Jan 2026 15 harvested: 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 5 where Syntology's instrument failed) · 6 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Step Potential Advantage Estimation: Harnessing Intermediate Confidence and... arXiv:2601.03823 7 Jan 2026 3 harvested: 2 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; 0 where Syntology's instrument failed) · 1 unverified refused by a rule: one configuration gave two values for one column
TokenSqueeze: Performance-Preserving Compression for Reasoning LLMs arXiv:2511.13223 17 Nov 2025 20 harvested: 15 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 1 honoured, 3 violated, 5 with no contract checked; 6 where Syntology's instrument failed) · 5 unverified placed
PuzzleMoE: Efficient Compression of Large Mixture-of-Experts Models via Sparse Expert... arXiv:2511.04805 6 Nov 2025 8 harvested: 2 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; 2 where Syntology's instrument failed) · 6 unverified refused by a rule: the table's column header does not name the leaderboard's metric
CGES: Confidence-Guided Early Stopping for Efficient and Accurate Self-Consistency arXiv:2511.02603 4 Nov 2025 19 harvested: 15 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 0 violated, 3 with no contract checked; 10 where Syntology's instrument failed) · 4 unverified rejected by the independent check
Reject Only Critical Tokens: Pivot-Aware Speculative Decoding arXiv:2511.00351 1 Nov 2025 16 harvested: 14 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 1 violated, 9 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified refused by a rule: the model gave no value to check against the cell
Teaching Language Models to Reason with Tools arXiv:2510.20342 23 Oct 2025 11 harvested: 11 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 2 violated, 3 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
EEPO: Exploration-Enhanced Policy Optimization via Sample-Then-Forget arXiv:2510.05837 7 Oct 2025 repository linked, no samples harvested refused by a rule: the table's column header does not name the leaderboard's metric
Beyond Two-Stage Training: Cooperative SFT and RL for LLM Reasoning arXiv:2509.06948 8 Sep 2025 repository linked, no samples harvested refused by a rule: the table's column header does not name the leaderboard's metric

Syntology 42 newer papers judged to report on this board: the papers counted above as having proposed a result (2 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 33 counted above as giving no cell that passed the extractor's own checks are the 33 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