Papers › QA-GNN: Reasoning with Language Models and Knowledge Graphs for Question Answering

QA-GNN: Reasoning with Language Models and Knowledge Graphs for Question Answering

13 Apr 2021NAACL 2021 4arXiv:2104.06378archive 2025-07-28

Michihiro Yasunaga, Hongyu Ren, Antoine Bosselut, Percy Liang, Jure Leskovec

The problem of answering questions using knowledge from pre-trained language models (LMs) and knowledge graphs (KGs) presents two challenges: given a QA context (question and answer choice), methods need to (i) identify relevant knowledge from large KGs, and (ii) perform joint reasoning over the QA context and KG. In this work, we propose a new model, QA-GNN, which addresses the above challenges through two key innovations: (i) relevance scoring, where we use LMs to estimate the importance of KG nodes relative to the given QA context, and (ii) joint reasoning, where we connect the QA context and KG to form a joint graph, and mutually update their representations through graph neural networks. We evaluate our model on QA benchmarks in the commonsense (CommonsenseQA, OpenBookQA) and biomedical (MedQA-USMLE) domains. QA-GNN outperforms existing LM and LM+KG models, and exhibits capabilities to perform interpretable and structured reasoning, e.g., correctly handling negation in questions.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 25 code samples harvested from 3 repositories linked to this paper: 8 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 17 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 21 samples from 1 repository, 8 ran (8 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 4 samples from 2 repositories, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

michiyasunaga/qagnn officialmentioned in papermentioned on GitHubpytorchMITSyntology: 21 samples, 8 ran (0 instrument-failed) · 13 unverified report
CMSC470-Team/Model mentioned on GitHubpytorchSyntology: not harvested report
HaochenLiu2000/QAP mentioned on GitHubpytorchSyntology: 2 samples, 0 ran · 2 unverified report
rucaibox/safe mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Great Truths are Always Simple: A Rather Simple Knowledge Encoder for Enhancing the... (arXiv:2205.01841; that paper's own run record: 0 ran · 4 unverified) and ACL Anthology id 2022.findings-naacl.131 (that paper's own run record: 13 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 1 honoured, 0 violated, 12 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
tobias-opsahl/fact-or-fiction mentioned on GitHubpytorchSyntology: 2 samples, 0 ran · 2 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

25 samples harvested: 8 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 17 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
7ran
17unverified

Licence: 5 of the 25 samples are pointer only, meaning Syntology does not serve that copy's text. For 1 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

michiyasunaga/qagnn official repository 21 samples, 8 ran (0 instrument-failed) · 13 unverified
create_hypothesis in 5 papers utils/convert_csqa.py ran fingerprinted MIT (permissive) · e4b000b919d28344 · report fixture: signature fitb:str/0/str, choice:str/0/str, ans_pos:bool/0/bool · returned str · level recorded 2026-09-25
create_output_dict utils/convert_obqa.py ran MIT (permissive) · 345456d568545c8e · report fixture: signature input_json:dict/0/?, statement:str/0/str, label:bool/0/bool · returned dict · level recorded 2026-09-25
del_pos in 6 papers utils/conceptnet.py ran fingerprinted MIT (permissive) · 5f4eefd99c11e698 · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-25
dropout_mask in 7 papers utils/layers.py ran MIT (permissive) · 866c784c24ba2fa8 · report fixture: signature x:Tensor/2/float, sz:tuple/1/int, p:float/0/float · returned Tensor · level recorded 2026-09-25
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
utils/layers.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · b75e9c793d2a286a · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
get_fitb_from_question in 6 papers utils/convert_csqa.py ran fingerprinted MIT (permissive) · a9a454bf4414e3fe · report fixture: signature question_text:str/0/str · returned str · level recorded 2026-09-25
load_cpnet_vocab utils/grounding.py ran MIT (permissive) · 2f104114280df441 · report fixture: signature cpnet_vocab_path:str/0/str · returned list · level recorded 2026-09-25
normalize_sparse_adj utils/maths.py ran MIT (permissive) · 5edf23a7957ba803 · report fixture: signature A:coo_matrix/2/int, sparse_type:str/0/str · returned coo_matrix · level recorded 2026-09-25
concepts2adj utils/graph.py unverified MIT (permissive) · e6a99b8687aab99c · report
construct_graph in 9 papers utils/conceptnet.py unverified MIT (permissive) · be8c35c9c1d52b66 · report
convert_qajson_to_entailment utils/convert_csqa.py unverified MIT (permissive) · 15baaf34a0e42ec6 · report
evaluate_accuracy in 17 papers qagnn.py unverified MIT (permissive) · 086fa0038784db8c · report
generate_adj_matrix_per_inst utils/graph.py unverified MIT (permissive) · 96ec8d520ae857f1 · report
glove_init in 5 papers utils/conceptnet.py unverified MIT (permissive) · 9dcf00b849de8977 · report
lemmatize utils/grounding.py unverified MIT (permissive) · 5ff1624579da2bdc · report
load_bert_xlnet_roberta_input_tensors utils/data_utils.py unverified MIT (permissive) · b5768cf5e27d534d · report
load_gpt_input_tensors utils/data_utils.py unverified MIT (permissive) · e633666b83ca73b6 · report
load_sparse_adj_data_with_contextnode utils/data_utils.py unverified MIT (permissive) · 16bd49eb2e3b8970 · report
make_one_hot in 15 papers modeling/modeling_qagnn.py unverified MIT (permissive) · df5e0b02401bf941 · report
plain_graph_generation utils/graph.py unverified MIT (permissive) · 66135965fe362f47 · report
HaochenLiu2000/QAP community (archive-listed) 2 samples, 0 ran · 2 unverified
CustomGraphTransformerLayer medqa/train_qap.py unverified no licence file found · pointer only · 2d38a4d448347c3c · report
GraphTransformer in 5 papers medqa/train_qap.py unverified no licence file found · pointer only · 941efc4033b96f8c · report
tobias-opsahl/fact-or-fiction community (archive-listed) 2 samples, 0 ran · 2 unverified
QAGNN models.py unverified no licence file found · pointer only · f82bdec01f779733 · report
get_bert_model models.py unverified no licence file found · pointer only · 216445b74badf787 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for QA-GNN (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for QA-GNN (graph method: QA-GNN). 2 bricks; individually on a fixture: 0 ran · 2 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. QAGNN entry · level 0 · unverified · f82bdec01f779733
  2. get_bert_model model · level 0 · unverified · 216445b74badf787

harvested Assembly for QA-GNN (graph method: QA-GNN). 2 bricks; individually on a fixture: 0 ran · 2 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. GraphTransformer entry · level 0 · unverified · 941efc4033b96f8c
  2. CustomGraphTransformerLayer model · level 0 · unverified · 2d38a4d448347c3c

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2104.06378") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Common Sense ReasoningGraph Representation LearningKnowledge GraphsLanguage ModellingMulti-hop Question AnsweringNegationQuestion AnsweringRiddle Sense

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Common Sense Reasoning CommonsenseQA QA-GNN Accuracy 76.1 #13 of 38 Archive leaderboard report
Question Answering OpenBookQA AristoRoBERTa + QA-GNN Accuracy 82.8 #20 of 45 Archive leaderboard report
Question Answering OpenBookQA QA-GNN Accuracy 82.8 #21 of 45 Archive leaderboard report
Question Answering OpenBookQA AristoRoBERTa Accuracy 77.8 #25 of 45 Archive leaderboard report
Riddle Sense RiddleSense QAGNN Accuracy (%) 67 #2 of 3 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGATLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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