SQuAD2.0 Benchmark (Question Answering)

286 rows 26 with code listed 2 metrics Dataset page

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

Metric Extra training data

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

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; 27 of 286 rows carry one.

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	IE-Net (ensemble)	90.939	93.214	–	–	–		not matched	report
2	FPNet (ensemble)	90.871	93.183	–	–	–		not matched	report
3	IE-NetV2 (ensemble)	90.860	93.100	–	–	–		not matched	report
4	SA-Net on Albert (ensemble)	90.724	93.011	–	–	–		not matched	report
5	SA-Net-V2 (ensemble)	90.679	92.948	–	–	–		not matched	report
6	FPNet (ensemble)	90.600	92.899	–	–	–		not matched	report
7	Retro-Reader (ensemble)	90.578	92.978	–	Paper	Code	2020	linked, not harvested	report
8	EntitySpanFocusV2 (ensemble)	90.521	92.824	–	–	–		not matched	report
9	TransNets + SFVerifier + SFEnsembler (ensemble)	90.487	92.894	–	–	–		not matched	report
10	EntitySpanFocus+AT (ensemble)	90.454	92.748	–	–	–		not matched	report
11	ATRLP+PV (ensemble)	90.442	92.877	–	–	–		not matched	report
12	LANetV2 (ensemble)	90.420	92.807	–	–	–		not matched	report
13	ELECTRA+ALBERT+EntitySpanFocus (ensemble)	90.420	92.799	–	–	–		not matched	report
14	ALBERT + DAAF + Verifier	90.386	92.777	–	–	–		not matched	report
15	electra+nlayers+adv+ds	90.284	92.691	–	–	–		not matched	report
16	MixEnsemble (ensemble)	90.194	92.594	–	–	–		not matched	report
17	electra+nlayers (ensemble)	90.126	92.535	–	–	–		not matched	report
18	Retro-Reader on ALBERT (ensemble)	90.115	92.580	–	–	–		not matched	report
19	Retro-Reader on ALBERT (ensemble)	90.115	92.580	–	Paper	Code	2020	linked, not harvested	report
20	ANet	90.081	92.457	–	–	–		not matched	report
21	Answer Dependent Classify (single model)	90.059	92.517	–	–	–		not matched	report
22	electra+nlayers+kdav (ensemble)	90.002	92.497	–	–	–		not matched	report
23	ALBERT + DAAF + Verifier (ensemble)	90.002	92.425	–	–	–		not matched	report
24	LANet (ensemble)	89.923	92.425	–	–	–		not matched	report
25	Electra-nlayers (ensemble)	89.777	92.312	–	–	–		not matched	report
26	Albert_Verifier_AA_Net (ensemble)	89.743	92.180	–	–	–		not matched	report
27	ALBERT (ensemble model)	89.731	92.215	–	Paper	Code	2019	46 of 126 ran · 80 unverified	report
28	Span Extract + Classify (single model)	89.562	92.226	–	–	–		not matched	report
29	Retro-Reader on ELECTRA (single model)	89.562	92.052	–	Paper	Code	2020	linked, not harvested	report
30	ELECTRA+ATRLP+PV (single model)	89.551	92.366	–	–	–		not matched	report
31	albert+transform+verify (ensemble)	89.528	92.059	–	–	–		not matched	report
32	albert+KD+transfer (ensemble)	89.461	92.134	–	–	–		not matched	report
33	ROaD-Electra (single model)	89.449	92.118	–	–	–		not matched	report
34	ROaD-Electra	89.449	92.118	–	–	–		not matched	report
35	Albert-nlayers (ensemble)	89.404	91.964	–	–	–		not matched	report
36	ELECTRA + E-Verifier (ensemble)	89.348	91.985	–	–	–		not matched	report
37	ELECTRA + ROBERTA + ALBERT (ensemble)	89.325	91.994	–	–	–		not matched	report
38	2task (single model)	89.325	91.939	–	–	–		not matched	report
39	Deberta	89.235	91.900	–	–	–		not matched	report
40	ALBERT + MTDA + SFVerifier (ensemble model)	89.235	91.739	–	–	–		not matched	report
41	ALBert-LSTM (ensemble)	89.224	91.853	–	–	–		not matched	report
42	ALBERT + SFVerifier (ensemble model)	89.133	91.666	–	–	–		not matched	report
43	ELECTRA+RL+EV (single model)	89.021	91.765	–	–	–		not matched	report
44	electra & albert (ensemble)	88.998	91.635	–	–	–		not matched	report
45	AE-TEST	88.998	91.635	–	–	–		not matched	report
46	ELECTRA+EntitySpanFocus (Single model)	88.874	91.546	–	–	–		not matched	report
47	SA-Net on Electra (single model)	88.851	91.486	–	–	–		not matched	report
48	ALBERT+Entailment DA (ensemble)	88.761	91.745	–	–	–		not matched	report
49	ELECTRA (single model)	88.716	91.365	–	–	–		not matched	report
50	Tuned ALBERT (ensemble model)	88.637	91.230	–	–	–		not matched	report
51	ELECTRA_ATT (single model)	88.614	91.303	–	–	–		not matched	report
52	Deberta+prefix	88.603	91.299	–	–	–		not matched	report
53	ALBERT (Single model)	88.592	91.286	–	–	–		not matched	report
54	XLNet + DAAF + Verifier (ensemble)	88.592	90.859	–	–	–		not matched	report
55	ALBERT + IG + NE (single model)	88.569	91.287	–	–	–		not matched	report
56	ALBERT + IG (single model)	88.524	91.256	–	–	–		not matched	report
57	aanet_v2.0 (single model)	88.434	90.918	–	–	–		not matched	report
58	albert+verifier (single model)	88.355	91.019	–	–	–		not matched	report
59	albert+KD+transfer (single)	88.298	91.078	–	–	–		not matched	report
60	UPM (ensemble)	88.231	90.713	–	–	–		not matched	report
61	ALBERT + SFVerifier (single model)	88.197	90.830	–	–	–		not matched	report
62	{alber_m_transfor} (single model)	88.186	90.939	–	–	–		not matched	report
63	XLNet + SG-Net Verifier (ensemble)	88.174	90.702	–	Paper	Code	2019	linked, not harvested	report
64	Retro-Reader on ALBERT (single model)	88.107	91.419	–	–	–		not matched	report
65	Retro-Reader on ALBERT (single model)	88.107	91.419	–	Paper	Code	2020	linked, not harvested	report
66	ALBERT (single model)	88.107	90.902	–	–	–		not matched	report
67	ALBERT (single model)	88.107	90.902	–	Paper	Code	2019	46 of 126 ran · 80 unverified	report
68	{MTL} (single model)	88.107	90.902	–	–	–		not matched	report
69	MTL (single model)	88.107	90.902	–	–	–		not matched	report
70	albert_with_tricks (single)	88.073	91.179	–	–	–		not matched	report
71	ALBert (single-model)	88.050	91.036	–	–	–		not matched	report
72	XLNet + SG-Net Verifier (ensemble)	88.050	90.645	–	–	–		not matched	report
73	DeBERTalarge	88.0	90.7	–	Paper	Code	2020	4 of 13 ran · 9 unverified	report
74	ALBERT 1.1 Th (single model)	87.994	90.944	–	–	–		not matched	report
75	SkERT-Large (single model)	87.994	90.944	–	–	–		not matched	report
76	albert+KD+transfer+twopass (single)	87.949	90.818	–	–	–		not matched	report
77	XLNet (single model)	87.926	90.689	–	Paper	Code	2019	10 of 24 ran · 14 unverified	report
78	ALBERT+RL (single model)	87.870	90.823	–	–	–		not matched	report
79	ALBERT+Entailment DA Verifier (single model)	87.847	91.265	–	–	–		not matched	report
80	Tuned ALBERT (single model)	87.847	90.532	–	–	–		not matched	report
81	albert_xxlarge (single model)	87.802	90.872	–	–	–		not matched	report
82	ALBERT 1.1 (single model)	87.700	90.588	–	–	–		not matched	report
83	LUKE (single model)	87.429	90.163	–	–	–		not matched	report
84	LUKE (single model)	87.429	90.163	–	Paper	Code	2020	3 of 10 ran · 7 unverified	report
85	XLNet + SG-Net Verifier++ (single model)	87.238	90.071	–	Paper	Code	2019	linked, not harvested	report
86	UPM (single model)	87.193	89.934	–	–	–		not matched	report
87	BERT + DAE + AoA (ensemble)	87.147	89.474	–	–	–		not matched	report
88	XLNet + SG-Net Verifier (single model)	87.046	89.899	–	–	–		not matched	report
89	RoBERTa+Verify (ensemble)	86.933	90.037	–	–	–		not matched	report
90	RoBERTa (single model)	86.820	89.795	–	Paper	Code	2019	22 of 48 ran · 26 unverified	report
91	RoBERTa (single model)	86.820	89.795	–	–	–		not matched	report
92	BERT + ConvLSTM + MTL + Verifier (ensemble)	86.730	89.286	–	–	–		not matched	report
93	BERT + N-Gram Masking + Synthetic Self-Training (ensemble)	86.673	89.147	–	–	–		not matched	report
94	RoBERTa+Span (ensemble)	86.651	89.595	–	–	–		not matched	report
95	Xlnet+Verifier	86.594	89.082	–	–	–		not matched	report
96	Xlnet+Verifier(single model)	86.572	89.063	–	–	–		not matched	report
97	Xlnet+Verifier (single model)	86.572	89.063	–	–	–		not matched	report
98	RoBERTa+Verify (single model)	86.448	89.586	–	–	–		not matched	report
99	XLNET-123 (single model)	86.436	89.086	–	–	–		not matched	report
100	XLNET-V2-123+ (single model)	86.403	89.148	–	–	–		not matched	report
101	XLNet (single model)	86.346	89.133	–	–	–		not matched	report
102	SG-Net (ensemble)	86.211	88.848	–	–	–		not matched	report
103	SG-Net (ensemble)	86.211	88.848	–	Paper	Code	2019	linked, not harvested	report
104	BERT++(ensemble)	86.166	88.886	–	–	–		not matched	report
105	SemBERT(ensemble)	86.166	88.886	–	–	–		not matched	report
106	SemBERT(ensemble)	86.166	88.886	–	Paper	Code	2019	4 of 12 ran · 8 unverified	report
107	SemBERT (ensemble)	86.166	88.886	–	Paper	Code	2019	4 of 12 ran · 8 unverified	report
108	Enhanced Albert+Verifier (ensemble)	86.098	89.634	–	–	–		not matched	report
109	BERT + DAE + AoA (single model)	85.884	88.621	–	–	–		not matched	report
110	RoBERTa+Span (single model)	85.872	88.989	–	–	–		not matched	report
111	RoBERTa-Large (ensemble model)	85.872	88.793	–	–	–		not matched	report
112	BNDVnet (ensemble model)	85.850	88.449	–	–	–		not matched	report
113	Unnamed submission by guo	85.838	88.921	–	–	–		not matched	report
114	BERTSP(single model)	85.838	88.921	–	–	–		not matched	report
115	BERTSP (single model)	85.838	88.921	–	–	–		not matched	report
116	Enhanced Albert+Verifier3 (ensemble)	85.827	89.778	–	–	–		not matched	report
117	SENSEFORTH + XLNet (single model)	85.827	88.699	–	–	–		not matched	report
118	SpanBERT (single model)	85.748	88.709	–	–	–		not matched	report
119	SemBERT (ensemble model)	85.703	88.400	–	–	–		not matched	report
120	SpanBERT	85.7	88.7	–	Paper	Code	2019	3 of 15 ran · 12 unverified	report
121	BertCNN (single model)	85.240	87.901	–	–	–		not matched	report
122	SG-Net (single model)	85.229	87.926	–	–	–		not matched	report
123	SG-Net (single model)	85.229	87.926	–	Paper	Code	2019	linked, not harvested	report
124	RoBERTa-Large (single model)	85.173	88.425	–	–	–		not matched	report
125	BERT + N-Gram Masking + Synthetic Self-Training (single model)	85.150	87.715	–	–	–		not matched	report
126	BERT + MMFT + ADA (ensemble)	85.082	87.615	–	–	–		not matched	report
127	BNDVnet (single model)	85.003	87.833	–	–	–		not matched	report
128	BERT + ConvLSTM + MTL + Verifier (single model)	84.924	88.204	–	–	–		not matched	report
129	Insight-baseline-BERT (single model)	84.834	87.644	–	–	–		not matched	report
130	BertSpan (ensemble)	84.823	87.489	–	–	–		not matched	report
131	SemBERT (single model)	84.800	87.864	–	–	–		not matched	report
132	SemBERT (single model)	84.800	87.864	–	Paper	Code	2019	4 of 12 ran · 8 unverified	report
133	Hanvon_model(single model)	84.721	87.117	–	–	–		not matched	report
134	Hanvon_model (single model)	84.721	87.117	–	–	–		not matched	report
135	xlnet	84.642	88.000	–	–	–		not matched	report
136	xlnet(single model)	84.642	88.000	–	–	–		not matched	report
137	xlnet (single model)	84.642	88.000	–	–	–		not matched	report
138	BERT++(single model)	84.620	87.625	–	–	–		not matched	report
139	BERT + Synthetic Self-Training (ensemble)	84.292	86.967	–	–	–		not matched	report
140	BERT + Multiple-CNN (ensemble)(Kyonggi University	84.202	86.767	–	–	–		not matched	report
141	BERT + Multiple-CNN (ensemble)	84.202	86.767	–	–	–		not matched	report
142	RoberTa Parallel Adapters Singl	84.123	87.013	–	–	–		not matched	report
143	RoberTa Parallel Adapters Single	84.123	87.013	–	–	–		not matched	report
144	RoberTa+Parallel+Adapters (single model)	84.123	87.013	–	–	–		not matched	report
145	SemNet (single model)	83.819	86.669	–	–	–		not matched	report
146	Tuned BERT-1seq Large Cased (single model)	83.751	86.594	–	–	–		not matched	report
147	BERT finetune baseline (ensemble)	83.536	86.096	–	–	–		not matched	report
148	SynNet (single model)	83.525	86.222	–	–	–		not matched	report
149	Lunet + Verifier + BERT (ensemble)	83.469	86.043	–	–	–		not matched	report
150	PAML+BERT (ensemble model)	83.457	86.122	–	–	–		not matched	report
151	Unnamed submission by Senseforth_AI	83.142	85.873	–	–	–		not matched	report
152	SENSEFORTH + BERT	83.142	85.873	–	–	–		not matched	report
153	Bert-raw (ensemble)	83.119	85.510	–	–	–		not matched	report
154	BERT with Something (ensemble)	83.051	85.737	–	–	–		not matched	report
155	BERT + MMFT + ADA (single model)	83.040	85.892	–	–	–		not matched	report
156	Lunet + Verifier + BERT (single model)	82.995	86.035	–	–	–		not matched	report
157	BERT + Synthetic Self-Training (single model)	82.972	85.810	–	–	–		not matched	report
158	mgrc_l (single model)	82.961	86.075	–	–	–		not matched	report
159	ATB (single model)	82.882	86.002	–	–	–		not matched	report
160	Tuned BERT Large Cased (single model)	82.803	85.863	–	–	–		not matched	report
161	BERT-Base + QA Pre-training (single model)	82.724	85.491	–	–	–		not matched	report
162	BERT + NeurQuRI (ensemble)	82.713	85.584	–	–	–		not matched	report
163	PAML+BERT (single model)	82.577	85.603	–	–	–		not matched	report
164	Unnamed submission by cooelf	82.431	85.178	–	–	–		not matched	report
165	AoA + DA + BERT (ensemble)	82.374	85.310	–	–	–		not matched	report
166	BART + Adapters + Lohfink-Rossi-Leaveout (single-model)	82.306	85.670	–	–	–		not matched	report
167	BERT finetune baseline (single model)	82.126	84.820	–	–	–		not matched	report
168	Candi-Net+BERT (ensemble)	82.126	84.624	–	–	–		not matched	report
169	BERT-Base PMI-Masking Additional Data (single model)	82.024	84.854	–	–	–		not matched	report
170	BERT_s (single model)	81.979	84.846	–	–	–		not matched	report
171	Unnamed submission by zw10	81.731	84.862	–	–	–		not matched	report
172	BERT-large+UBFT (single model)	81.573	84.535	–	–	–		not matched	report
173	AoA + DA + BERT (single model)	81.178	84.251	–	–	–		not matched	report
174	BERT with Something (single model)	81.110	84.386	–	–	–		not matched	report
175	BERT-Base PMI-Masking (single model)	80.896	83.604	–	–	–		not matched	report
176	BERT + UnAnsQ (single model)	80.749	83.851	–	–	–		not matched	report
177	BERT + AL (single model)	80.715	83.827	–	–	–		not matched	report
178	BERT + NeurQuRI (single model)	80.591	83.391	–	–	–		not matched	report
179	BERTlarge (ensemble)	80.456	83.509	–	–	–		not matched	report
180	Bert	80.422	83.118	–	–	–		not matched	report
181	Bert-raw (single)	80.411	83.457	–	–	–		not matched	report
182	Candi-Net+BERT (single model)	80.388	82.908	–	–	–		not matched	report
183	PMI-Masking Additional Data Random Baseline (single model)	80.377	83.262	–	–	–		not matched	report
184	Unnamed submission by zw4	80.354	83.329	–	–	–		not matched	report
185	{Bert-Span} (single model)	80.354	83.329	–	–	–		not matched	report
186	Bert-raw (single model)	80.343	83.243	–	–	–		not matched	report
187	PMI-Masking Pure-PMI (single model)	80.241	83.175	–	–	–		not matched	report
188	BISAN-CC (single model)	80.208	83.149	–	–	–		not matched	report
189	ST_bl	80.140	82.962	–	–	–		not matched	report
190	PwP+BERT (single model)	80.117	83.189	–	–	–		not matched	report
191	PMI-Masking Random Baseline (single model)	80.038	82.796	–	–	–		not matched	report
192	BERT + UDA (single model)	80.005	83.208	–	–	–		not matched	report
193	PMI-Masking Additional Data Pure-PMI (single model)	79.993	83.039	–	–	–		not matched	report
194	Original BERT Large Cased (single model)	79.971	83.266	–	–	–		not matched	report
195	bert (single model)	79.971	83.184	–	–	–		not matched	report
196	BERT + Sparse-Transformer	79.948	83.023	–	–	–		not matched	report
197	Insight-baseline (single model)	79.779	83.099	–	–	–		not matched	report
198	NEXYS_BASE (single model)	79.779	82.912	–	–	–		not matched	report
199	BERT uncased (single model)	79.745	83.020	–	–	–		not matched	report
200	{bert-finetuning} (single model)	79.632	82.852	–	–	–		not matched	report
201	L6Net + BERT (single model)	79.181	82.259	–	–	–		not matched	report
202	Fusion Adapters TriviaQA NQ Singl	78.933	81.863	–	–	–		not matched	report
203	RoberTa Fusion Adapters Single	78.933	81.863	–	–	–		not matched	report
204	RoberTa+Fusion+Adapters (single model)	78.933	81.863	–	–	–		not matched	report
205	{Anonymous} (single model)	78.876	82.524	–	–	–		not matched	report
206	BERT + WIAN (ensemble)	78.650	81.497	–	–	–		not matched	report
207	BERTlarge (single model)	78.650	81.474	–	–	–		not matched	report
208	AMBERT (single model)	78.594	81.445	–	–	–		not matched	report
209	BISAN (single model)	78.481	81.531	–	–	–		not matched	report
210	BERT-Large-Cased (single model)	78.357	81.500	–	–	–		not matched	report
211	BERT-Large-Cased	78.357	81.500	–	–	–		not matched	report
212	BERT+AC(single model)	78.052	81.174	–	–	–		not matched	report
213	BERT+AC (single model)	78.052	81.174	–	–	–		not matched	report
214	BERT (single model)	77.319	80.310	–	–	–		not matched	report
215	TriviaQA Adapter Single Tune	77.262	80.258	–	–	–		not matched	report
216	RoberTa Adapter Single	77.262	80.258	–	–	–		not matched	report
217	RoberTa+Adapter (single model)	77.262	80.258	–	–	–		not matched	report
218	SLQA+BERT (single model)	77.003	80.209	–	–	–		not matched	report
219	AMBERT-H (single model)	76.710	79.659	–	–	–		not matched	report
220	AMBERT-S (single model)	76.563	79.776	–	–	–		not matched	report
221	synss (single model )	76.055	79.329	–	–	–		not matched	report
222	synss (single model)	76.055	79.329	–	–	–		not matched	report
223	BERT-Base-L (single model)	75.457	78.232	–	–	–		not matched	report
224	mgrc	75.344	78.381	–	–	–		not matched	report
225	BERT-Base-V (single model)	75.073	77.805	–	–	–		not matched	report
226	MIR-MRC(F-Net) (single model)	74.791	77.988	–	–	–		not matched	report
227	MIR-MRC (F-Net)	74.791	77.988	–	–	–		not matched	report
228	BERT-Base-DT (single model)	74.769	77.706	–	–	–		not matched	report
229	ARSG-BERT (single model)	74.746	78.227	–	–	–		not matched	report
230	BERT-Base-V2	74.656	77.404	–	–	–		not matched	report
231	BERT-Base-DP (single model)	74.577	77.464	–	–	–		not matched	report
232	{BERTcw} (single model)	74.385	77.308	–	–	–		not matched	report
233	BERT-Base-Add (single model)	74.329	77.396	–	–	–		not matched	report
234	nlnet (single model)	74.272	77.052	–	–	–		not matched	report
235	batch2 (single model)	73.742	76.858	–	–	–		not matched	report
236	MMIPN	73.505	76.424	–	–	–		not matched	report
237	BERT-Base-Baseline (single model)	73.302	76.284	–	–	–		not matched	report
238	BERT-Base (single model)	73.099	76.236	–	–	–		not matched	report
239	ICL_MODEL(ensemble)	72.884	76.217	–	–	–		not matched	report
240	ICL_MODEL (ensemble)	72.884	76.217	–	–	–		not matched	report
241	YARCS (ensemble)	72.670	75.507	–	–	–		not matched	report
242	BERT-base	72.072	75.513	–	–	–		not matched	report
243	BERTBase (single model)	72.072	75.513	–	–	–		not matched	report
244	Reinforced Mnemonic Reader + Answer Verifier (single model)	71.767	74.295	–	Paper	–	2018	no code linked	report
245	BERT-Base (single)	71.699	74.430	–	–	–		not matched	report
246	BERT+Answer Verifier (single model)	71.666	75.457	–	–	–		not matched	report
247	SLQA+ (single model)	71.462	74.434	–	–	–		not matched	report
248	Unet (ensemble)	71.417	74.869	–	Paper	Code	2018	linked, not harvested	report
249	SAN (ensemble model)	71.316	73.704	–	Paper	Code	2017	linked, not harvested	report
250	HYDRA_BERT (single model)	71.293	74.578	–	–	–		not matched	report
251	{BERT-base} (single-model)	70.763	74.449	–	–	–		not matched	report
252	FusionNet++ (ensemble)	70.300	72.484	–	Paper	Code	2017	linked, not harvested	report
253	Multi-Level Attention Fusion(MLAF) (single model)	69.476	72.857	–	–	–		not matched	report
254	Multi-Level Attention Fusion (MLAF)	69.476	72.857	–	–	–		not matched	report
255	Unet (single model)	69.262	72.642	–	–	–		not matched	report
256	DocQA + NeurQuRI (single model)	68.766	71.662	–	–	–		not matched	report
257	SAN (single model)	68.653	71.439	–	Paper	Code	2017	linked, not harvested	report
258	KACTEIL-MRC(GFN-Net) (single model)	68.213	70.878	–	–	–		not matched	report
259	KACTEIL-MRC (GFN-Net)	68.213	70.878	–	–	–		not matched	report
260	BiDAF++ with pair2vec (single model)	68.021	71.583	–	–	–		not matched	report
261	VS^3-NET (single model)	67.897	70.884	–	–	–		not matched	report
262	EBB-Net (single model)	66.610	70.303	–	–	–		not matched	report
263	KakaoNet2 (single model)	65.719	69.381	–	–	–		not matched	report
264	BiDAF++ (single model)	65.651	68.866	–	–	–		not matched	report
265	abcNet (single model)	65.256	69.206	–	–	–		not matched	report
266	BiDAF + Self Attention + ELMo (single model)	63.372	66.251	–	Paper	Code	2018	23 of 58 ran · 35 unverified	report
267	BiDAF + Self Attention + ELMo (single model)	63.372	66.251	–	–	–		not matched	report
268	BSAE AddText (single model)	63.338	67.422	–	–	–		not matched	report
269	eeAttNet (single model)	63.327	66.633	–	–	–		not matched	report
270	BiDAF + Self Attention (single model)	59.332	62.305	–	–	–		not matched	report
271	BiDAF-No-Answer (single model)	59.174	62.093	–	–	–		not matched	report
272	BNA + SoftDrop (single model)	58.508	62.045	–	–	–		not matched	report
273	Tree-LSTM + BiDAF + ELMo (single model)	57.707	62.341	–	–	–		not matched	report
274	BNA + HardDrop (single model)	56.545	59.546	–	–	–		not matched	report
275	Unnamed submission by Simon	56.545	59.546	–	–	–		not matched	report
276	Anonymous (single model)	49.695	49.701	–	–	–		not matched	report
277	{FOO} (single model)	48.883	48.883	–	–	–		not matched	report
278	Bert Large Sentence (Single Model)	48.804	48.815	–	–	–		not matched	report
279	XLNet + DAAF + BERTverifier (ensemble)	44.945	47.994	–	–	–		not matched	report
280	TSAN	40.397	43.213	–	–	–		not matched	report
281		27.217	29.597	–	–	–		not matched	report
282	CHECK SYSTEM (single model)	4.830	5.920	–	–	–		not matched	report
283	5cls_squad1_fai	0.068	3.971	–	–	–		not matched	report
284	LUKE 483M		90.2	–	Paper	Code	2020	3 of 10 ran · 7 unverified	report
285	Ensemble ALBERT		90.123	–	Paper	Code	2021	linked, not harvested	report
286	gMLP-large		78.3	–	Paper	Code	2021	34 of 44 ran · 10 unverified	report

All 286 rows shown. 27 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 of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded 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). 13 rows have a graph line, from 9 distinct papers; 13 rows (9 papers) have at least one sample that ran. Counting each paper once: Syntology ran 149 of 350 samples; 201 unverified. Separately, 91 of those 350 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. Read from the graph 2026-09-24. Per-sample status is on the paper page.

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