Browse › Natural Language Processing › Question Answering › TriviaQA

Question Answering archive 2025-07-28

TriviaQA Benchmark (Question Answering)

56 rows 36 with code listed 2 metrics Dataset page

Syntology Since the archive (Syntology): none placed; 20 newer papers judged to report on this board so far.

Question answering can be segmented into domain-specific tasks like community question answering and knowledge-base question answering. Popular benchmark datasets for evaluation question answering systems include SQuAD, HotPotQA, bAbI, TriviaQA, WikiQA, and many others. Models for question answering are typically evaluated on metrics like EM and F1. Some recent top performing models are T5 and XLNet.

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

Over time archive 2025-07-28

Direction inferred from the metric name, not from the archive: F1 (higher is better). Not inferred (points only, no best-so-far line): EM. Points are placed at the row's paper date, not an evaluation date; 55 of 56 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 Claude 2 (few-shot, k=5) 87.5 – Paper – 2023 no code linked report
2 GPT-4-0613 87 – – – not matched report
3 Claude 1.3 (few-shot, k=5) 86.7 – Paper – 2023 no code linked report
4 RankRAG-llama3-70b (Zero-Shot, KILT) 86.5 ✓ Paper – 2024 no code linked report
5 PaLM 2-L (one-shot) 86.1 ✓ Paper Code 2023 linked, not harvested report
6 ChatQA-1.5-llama3-70b (Zero-Shot, KILT) 85.6 ✓ Paper – 2024 no code linked report
7 LLaMA 2 70B (one-shot) 85 – Paper Code 2023 community repositories only · 33 ran (of which 7 constructed an object rather than computing a result; 20 with no instrument failure: 1 honoured, 1 violated, 18 with no contract checked; 13 where Syntology's instrument failed) · 19 unverified report
8 GPT-4-0613 (Zero-shot) 84.8 – 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
9 RankRAG-llama3-8b (Zero-Shot, KILT) 82.9 ✓ Paper – 2024 no code linked report
10 PaLM 2-M (one-shot) 81.7 – Paper Code 2023 linked, not harvested report
11 PaLM-540B (Few-Shot) 81.4 ✓ Paper Code 2022 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 report
12 PaLM-540B (One-Shot) 81.4 – Paper Code 2022 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 report
13 ChatQA-1.5-llama3-8B (Zero-Shot, KILT) 81.0 ✓ Paper – 2024 no code linked report
14 GaC(Qwen2-72B-Instruct + Llama-3-70B-Instruct) 79.29 – Paper Code 2024 official (archive's flag): 5 ran · 5 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; 3 where Syntology's instrument failed) · 2 unverified report
15 Claude Instant 1.1 (few-shot, k=5) 78.9 – Paper – 2023 no code linked report
16 code-davinci-002 175B + REPLUG LSR (Few-Shot) 77.3 – Paper Code 2023 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) · 3 unverified report
17 PaLM-540B (Zero-Shot) 76.9 – Paper Code 2022 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 report
18 code-davinci-002 175B + REPLUG (Few-Shot) 76.8 – Paper Code 2023 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) · 3 unverified report
19 GLaM 62B/64E (One-shot) 75.8 ✓ Paper – 2021 no code linked report
20 GLaM 62B/64E (Few-shot) 75.8 – Paper – 2021 no code linked report
21 RA-DIT (Zero-Shot) 75.4 ✓ Paper – 2023 no code linked report
22 PaLM 2-S (one-shot) 75.2 – Paper Code 2023 linked, not harvested report
23 LLaMA 65B (few-shot, k=64) 73.0 – 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
24 FiE+PAQ 72.6 – Paper – 2022 no code linked report
25 LLaMA 65B (few-shot, k=5) 72.6 – 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
26 RankRAG-llama3-70b (Zero-Shot, DPR) 72.6 ✓ Paper – 2024 no code linked report
27 FiD+Distil 72.1 ✓ Paper Code 2020 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) · 0 unverified report
28 LLaMA 65B (one-shot) 71.6 – 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
29 EMDR2 71.4 – Paper Code 2021 official (archive's flag): 2 ran · 4 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; 4 where Syntology's instrument failed) · 0 unverified report
30 GLaM 62B/64E (Zero-shot) 71.3 – Paper – 2021 no code linked report
31 GPT-3 175B (Few-Shot) 71.2 ✓ 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 Mistral 7B (5-shot) 69.9 – 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
33 ChatQA-1.5-llama3-70b (Zero-Shot, DPR) 69.0 – Paper – 2024 no code linked report
34 LLaMA 65B (zero-shot) 68.2 – 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
35 Fusion-in-Decoder (large) 67.6 – Paper Code 2020 linked, not harvested report
36 MemoReader 67.2173.26 ✓ Paper – 2018 no code linked report
37 S-Norm 66.3771.32 ✓ Paper Code 2017 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) · 2 unverified report
38 TOME-2 65.8 – Paper Code 2021 linked, not harvested report
39 Shakti-LLM (2.5B) 58.2 – Paper – 2024 no code linked report
40 Branch-Train-MiX 4x7B (sampling top-2 experts) 57.1 – Paper Code 2024 linked, not harvested report
41 DPR 56.8 – Paper Code 2020 official: harvested, nothing ran · 11 ran (of which 6 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified report
42 FLAN 137B (zero-shot) 56.7 – Paper Code 2021 official: harvested for another paper · 0 ran · 1 unverified report
43 RAG 56.1 – Paper Code 2020 6 ran (of which 3 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified report
44 Reading Twice for NLU 50.5656.73 – Paper – 2017 no code linked report
45 Mnemonic Reader 46.9452.85 – Paper Code 2017 linked, not harvested report
46 ORQA 45 – Paper Code 2019 linked, not harvested report
47 MEMEN 43.1646.90 – Paper – 2017 no code linked report
48 SpanBERT 83.6 – Paper Code 2019 official (archive's flag): 2 ran · 9 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 6 unverified report
49 BigBird-etc 80.9 – Paper Code 2020 official (archive's flag): 1 ran · 10 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 8 where Syntology's instrument failed) · 5 unverified report
50 DPA-RAG 80.1 – Paper Code 2024 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) · 2 unverified report
51 LinkBERT (large) 78.2 – Paper Code 2022 official: harvested, nothing ran · 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 11 unverified report
52 DyREX 77.37 – Paper Code 2022 official (archive's flag): 1 ran · 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) · 1 unverified report
53 Search-o1 74.1 – Paper Code 2025 linked, not harvested report
54 UnitedQA (Hybrid reader) 70.3 – Paper – 2021 no code linked report
55 ReasonBERTR 45.5 – Paper Code 2021 official (archive's flag): 9 ran · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified report
56 ReasonBERTB 37.2 – Paper Code 2021 official (archive's flag): 9 ran · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified report

All 56 rows shown. 55 link to a paper page on this site; 13 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). 27 rows have a graph line, from 20 distinct papers. Rows whose paper has a sample that ran: 26 (23 with a run with no instrument failure, 3 where every run was a failure of Syntology's instrument). Papers: 19 with a sample that ran: 16 with a run with no instrument failure, 3 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 27 rows with a graph line, 21 have an archive-official repository: on 10 a sample from it ran (6 with a run with no instrument failure), on 4 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 339 samples: 235 ran (of which 42 constructed an object rather than computing a result; 168 with no instrument failure: 12 honoured, 7 violated, 149 with no contract checked; 67 where Syntology's instrument failed) · 104 unverified. Separately, 85 of those 339 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), 40 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 1,580 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and 20 proposed a result; of those 20, 18 gave no cell that passed the extractor's own checks, 2 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
Look Before You Leap: Factual Decoding with Internal Attribution Signals arXiv:2609.15745 14 Sep 2026 repository linked, no samples harvested refused by a rule: the row is an ablation variant, not the method itself
Feedback-Assisted Trust Propagation over Document Relation Graphs for... arXiv:2609.00543 1 Sep 2026 12 harvested: 11 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; 5 where Syntology's instrument failed) · 1 unverified refused by a rule: the value is not on the scale of the leaderboard column's archive values
Prefilling-dLLM: Predictive Prefilling for Long-Context Inference in Diffusion Language Models arXiv:2606.10537 9 Jun 2026 19 harvested: 14 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; 9 where Syntology's instrument failed) · 5 unverified refused by a rule: the table's column header does not name the leaderboard's metric
From Context-Aware to Conflict-Aware: Generalizing Contrastive Decoding for Knowledge... arXiv:2606.10298 9 Jun 2026 2 harvested: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; the one sample that ran constructed an object rather than computing a result refused by a rule: several configurations of the method, and no metric direction to pick one
Efficient Agentic Reinforcement Learning with On-Policy Intrinsic Knowledge Boundary... arXiv:2605.26952 26 May 2026 9 harvested: 7 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; 5 where Syntology's instrument failed) · 2 unverified refused by a rule: one configuration gave two values for one column
Trust or Abstain? A Self-Aware RAG Approach arXiv:2605.18792 11 May 2026 held, no recorded run refused by a rule: the table's column header does not name the leaderboard's metric
Self-Induced Outcome Potential: Turn-Level Credit Assignment for Agents without Verifiers arXiv:2605.04984 6 May 2026 2 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) · 1 unverified refused by a rule: the value is not on the scale of the leaderboard column's archive values
Negative Advantages Is a Double-Edged Sword: Calibrating advantages in GRPO for Search Agents arXiv:2604.18235 20 Apr 2026 2 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) · 0 unverified refused by a rule: one configuration gave two values for one column
CoSearch: Joint Training of Reasoning and Document Ranking via Reinforcement Learning... arXiv:2604.17555 19 Apr 2026 7 harvested: 7 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; 5 where Syntology's instrument failed) · 0 unverified refused by a rule: the value is not on the scale of the leaderboard column's archive values
Skill-RAG: Failure-State-Aware Retrieval Augmentation via Hidden-State Probing and... arXiv:2604.15771 17 Apr 2026 13 harvested: 12 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 7 where Syntology's instrument failed) · 1 unverified refused by a rule: one configuration gave two values for one column
TIPS: Turn-Level Information-Potential Reward Shaping for Search-Augmented LLMs arXiv:2603.22293 11 Mar 2026 11 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) · 6 unverified refused by rules: the number the model quoted is not the number printed in the cell it pointed at; one configuration gave two values for one column
Rethinking Retrieval-Augmented Generation as a Cooperative Decision-Making Problem arXiv:2602.18734 21 Feb 2026 held, no recorded run refused by a rule: the table's column header does not name the leaderboard's metric
GLM-5: from Vibe Coding to Agentic Engineering arXiv:2602.15763 17 Feb 2026 repository linked, no samples harvested rejected by the independent check
Fix the Structural Bottleneck: Context Compression via Explicit Information Transmission arXiv:2602.03784 3 Feb 2026 5 harvested: 1 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; 0 where Syntology's instrument failed) · 4 unverified refused by a rule: one configuration gave two values for one column
Rethinking the Reranker: Boundary-Aware Evidence Selection for Robust... arXiv:2602.03689 3 Feb 2026 26 harvested: 24 ran (of which 4 constructed an object rather than computing a result; 10 with no instrument failure: 6 honoured, 0 violated, 4 with no contract checked; 14 where Syntology's instrument failed) · 2 unverified refused by a rule: one configuration gave two values for one column
Reasoning While Asking: Transforming Reasoning Large Language Models from Passive... arXiv:2601.22139 29 Jan 2026 12 harvested: 12 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 9 where Syntology's instrument failed) · 0 unverified refused by a rule: several configurations of the method, and no metric direction to pick one
ArcAligner: Adaptive Recursive Aligner for Compressed Context Embeddings in RAG arXiv:2601.05038 8 Jan 2026 repository linked, no samples harvested rejected by the independent check
EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle arXiv:2510.16079 17 Oct 2025 5 harvested: 5 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; 2 where Syntology's instrument failed) · 0 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Mamba Modulation: On the Length Generalization of Mamba arXiv:2509.19633 23 Sep 2025 12 harvested: 11 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 2 honoured, 1 violated, 3 with no contract checked; 5 where Syntology's instrument failed) · 1 unverified refused by a rule: the table's column header does not name the leaderboard's metric
PruneCD: Contrasting Pruned Self Model to Improve Decoding Factuality arXiv:2509.16598 20 Sep 2025 17 harvested: 14 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 9 where Syntology's instrument failed) · 3 unverified refused by a rule: one configuration gave two values for one column

Syntology 20 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 18 counted above as giving no cell that passed the extractor's own checks are the 18 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