Browse › Natural Language Processing › Question Answering › TruthfulQA

Question Answering archive 2025-07-28

TruthfulQA Benchmark (Question Answering)

33 rows 29 with code listed 10 metrics Dataset page

Syntology Since the archive (Syntology): 1 placed; 22 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: ROUGE (higher is better), BLEU (higher is better), Accuracy (higher is better). Not inferred (points only, no best-so-far line): MC1, MC2, % true, % info, % true (GPT-judge), BLEURT, EM. Points are placed at the row's paper date, not an evaluation date; 30 of 33 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 (RLHF) 0.59 – 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 Mistral-7B-Instruct-v0.2 + TruthX 0.560.75 – Paper Code 2024 official (archive's flag): 2 ran · 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) · 1 unverified; every one of the 2 samples that ran constructed an object rather than computing a result report
3 LLaMa-2-7B-Chat + TruthX 0.540.74 – Paper Code 2024 official (archive's flag): 2 ran · 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) · 1 unverified; every one of the 2 samples that ran constructed an object rather than computing a result report
4 LLaMA-2-Chat-13B + Representation Control (Contrast Vector) 0.54 – Paper Code 2023 official (archive's flag): 6 ran · 6 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; 2 where Syntology's instrument failed) · 4 unverified report
5 LLaMA-2-Chat-7B + Representation Control (Contrast Vector) 0.48 – Paper Code 2023 official (archive's flag): 6 ran · 6 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; 2 where Syntology's instrument failed) · 4 unverified report
6 Vicuna 7B + Inference Time Intervention (ITI) 0.38988.683.5 – – – not matched report
7 Alpaca 7B + Inference Time Intervention (ITI) 0.31966.697.7 – – – not matched report
8 Gopher 280B (zero-shot, Our Prompt + Choices) 0.295 – Paper Code 2021 linked, not harvested report
9 LLaMA 7B + Inference Time Intervention (ITI) 0.28845.193.8 – – – not matched report
10 GAL 120B 0.26 – 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
11 Gopher 7.1 (zero-shot, QA prompts) 0.25 – Paper Code 2021 linked, not harvested report
12 GAL 30B 0.24 – 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
13 Gopher 7.1B (zero-shot, Our Prompt + Choices) 0.23 – Paper Code 2021 linked, not harvested report
14 Gopher 1.4 (zero-shot, QA prompts) 0.23 – Paper Code 2021 linked, not harvested report
15 GPT-2 1.5B 0.220.3929.5089.8429.87-0.25-9.41-4.91 – Paper Code 2021 official (archive's flag): 2 ran · 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 report
16 Gopher 1.4B (zero-shot, Our Prompt + Choices) 0.217 – Paper Code 2021 linked, not harvested report
17 GPT-3 175B 0.210.3320.4497.5520.56-0.56-17.75-17.38 – Paper Code 2021 official (archive's flag): 2 ran · 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 report
18 OPT 175B 0.21 – 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
19 GPT-J 6B 0.200.3626.6889.9627.17-0.31-11.35-7.58 – Paper Code 2021 official (archive's flag): 2 ran · 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 report
20 UnifiedQA 3B 0.190.3553.8664.5053.240.081.76-0.16 – Paper Code 2021 official (archive's flag): 2 ran · 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 report
21 GAL 125M 0.19 – 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
22 GAL 1.3B 0.19 – 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
23 GAL 6.7B 0.19 – 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
24 Gopher 280B (zero-shot, QA prompts) 0. 27 – Paper Code 2021 linked, not harvested report
25 LLaMA 65B 5753 – 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 LLaMA 33B 5248 – 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
27 LLaMA 13B 4741 – 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 LLaMA 7B 3329 – 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 CoA 67.3 – Paper Code 2024 official (archive's flag): 8 ran · 8 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; 2 where Syntology's instrument failed) · 2 unverified report
30 ToT 66.6 – Paper Code 2023 official (archive's flag): 2 ran · 19 ran (of which 6 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified report
31 CoA w/o actions 63.3 – Paper Code 2024 official (archive's flag): 8 ran · 8 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; 2 where Syntology's instrument failed) · 2 unverified report
32 Auto-CoT 42.2 – Paper Code 2022 official: no sample here; runs from other or unrecorded repositories · 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
33 Shakti-LLM (2.5B) 68.4 – Paper – 2024 no code linked report

All 33 rows shown. 30 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). 23 rows have a graph line, from 9 distinct papers. Rows whose paper has a sample that ran: 23 (18 with a run with no instrument failure, 5 where every run was a failure of Syntology's instrument). Papers: 9 with a sample that ran: 7 with a run with no instrument failure, 2 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 23 rows with a graph line, 23 have an archive-official repository: on 17 a sample from it ran (13 with a run with no instrument failure), on 1 the samples that ran came from community repositories only, and on 5 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 117 samples: 82 ran (of which 17 constructed an object rather than computing a result; 61 with no instrument failure: 5 honoured, 3 violated, 53 with no contract checked; 21 where Syntology's instrument failed) · 35 unverified. Separately, 12 of those 117 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), 61 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 1,579 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and 22 proposed a result; of those 22, 20 gave no cell that passed the extractor's own checks, 1 was rejected by the independent check, 0 await a verdict and 1 was 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) MC2 Date Where in the paper Code Syntology Report
ARCHead: Activation-Metric Residual Correction for Large Language Model Output Heads arXiv:2608.02703 ARCHead 0.5408 3 Aug 2026 Table 10, row “ARCHead”Caption: “Table 10: Downstream accuracy sanity check. Small differences should not be…” 20 harvested: 19 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 16 where Syntology's instrument failed) · 1 unverified report

Syntology 1 entry, 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
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
IDEEA: training-free Input-Dependent stEEring via Activation cluster matching arXiv:2609.02089 2 Sep 2026 15 harvested: 13 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; 6 where Syntology's instrument failed) · 2 unverified refused by a rule: the number the model quoted is not the number printed in the cell it pointed at
CyrillicQA: The Influence of Phonetically Encoded Secret Language on LLM Performance arXiv:2608.21462 20 Aug 2026 repository linked, no samples harvested refused by a rule: the model gave no value to check against the cell
ARCHead: Activation-Metric Residual Correction for Large Language Model Output Heads arXiv:2608.02703 3 Aug 2026 20 harvested: 19 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 16 where Syntology's instrument failed) · 1 unverified placed
PEAR: Permutation-Equivariant Adaptive Routing Multi-Agent Debate arXiv:2606.20621 26 May 2026 7 harvested: 6 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; 3 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
Adaptive Steering and Remasking for Safe Generation in Diffusion Language Models arXiv:2605.13043 13 May 2026 9 harvested: 9 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; 7 where Syntology's instrument failed) · 0 unverified refused by a rule: the row is an ablation variant, not the method itself
FineSteer: A Unified Framework for Fine-Grained Inference-Time Steering in Large... arXiv:2604.15488 16 Apr 2026 held, no recorded run refused by a rule: the row is an ablation variant, not the method itself
Awakening Dormant Experts:Counterfactual Routing to Mitigate MoE Hallucinations arXiv:2604.14246 15 Apr 2026 repository linked, no samples harvested refused by rules: the value is not on the scale of the leaderboard column's archive values; the table reports a different split or setting than the leaderboard
DeCoVec: Building Decoding Space based Task Vector for Large Language Models via... arXiv:2604.11129 13 Apr 2026 1 harvested: 1 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; 0 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
DSPA: Dynamic SAE Steering for Data-Efficient Preference Alignment arXiv:2603.21461 23 Mar 2026 7 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) · 2 unverified refused by a rule: the value is not on the scale of the leaderboard column's archive values
Towards Self-Robust LLMs: Intrinsic Prompt Noise Resistance via CoIPO arXiv:2603.03314 9 Feb 2026 5 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) · 3 unverified refused by a rule: several configurations of the method, and no metric direction to pick one
Token-level Data Selection for Safe LLM Fine-tuning arXiv:2603.01185 1 Mar 2026 3 harvested: 3 ran (of which 1 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) · 0 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Accelerating LLM Pre-Training through Flat-Direction Dynamics Enhancement arXiv:2602.22681 26 Feb 2026 7 harvested: 6 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; 3 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
Faithful Bi-Directional Model Steering via Distribution Matching and Distributed... arXiv:2602.05234 5 Feb 2026 2 harvested: 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 no cell passed the extractor's checks
Variance-Adaptive Muon: Accelerating LLM Pretraining with NSR-Modulated and... arXiv:2601.14603 21 Jan 2026 7 harvested: 5 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; 3 where Syntology's instrument failed) · 2 unverified refused by a rule: the table's column header does not name the leaderboard's metric
Rendering Data Unlearnable by Exploiting LLM Alignment Mechanisms arXiv:2601.03401 6 Jan 2026 8 harvested: 7 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; 0 where Syntology's instrument failed) · 1 unverified rejected by the independent check
MeCeFO: Enhancing LLM Training Robustness via Fault-Tolerant Optimization arXiv:2510.16415 18 Oct 2025 7 harvested: 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) · 3 unverified refused by a rule: several configurations of the method, and no metric direction to pick one
The Alignment Waltz: Jointly Training Agents to Collaborate for Safety arXiv:2510.08240 9 Oct 2025 repository linked, no samples harvested refused by a rule: the row is an ablation variant, not the method itself
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 rules: the value is not on the scale of the leaderboard column's archive values; the table's column header does not name the leaderboard's metric; one configuration gave two values for one column
Distribution-Aligned Decoding for Efficient LLM Task Adaptation arXiv:2509.15888 19 Sep 2025 1 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) · 0 unverified; the one sample that ran constructed an object rather than computing a result refused by rules: the value is not on the scale of the leaderboard column's archive values; the quoted value appears more than once in the row; the table's column header does not name the leaderboard's metric; one configuration gave two values for one column
ConfTuner: Training Large Language Models to Express Their Confidence Verbally arXiv:2508.18847 26 Aug 2025 4 harvested: 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 refused by a rule: the value is not on the scale of the leaderboard column's archive values
Steering When Necessary: Flexible Steering Large Language Models with Backtracking arXiv:2508.17621 25 Aug 2025 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 rules: the value is not on the scale of the leaderboard column's archive values; several configurations of the method, and no metric direction to pick one

Syntology 22 newer papers judged to report on this board: the papers counted above as having proposed a result (1 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 20 counted above as giving no cell that passed the extractor's own checks are the 19 marked “refused by a rule” and the 1 marked “no cell passed the extractor's checks” together. 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