Papers › Robust Retrieval Augmented Generation for Zero-shot Slot Filling

Robust Retrieval Augmented Generation for Zero-shot Slot Filling

31 Aug 2021EMNLP 2021 11arXiv:2108.13934archive 2025-07-28

Michael Glass, Gaetano Rossiello, Md Faisal Mahbub Chowdhury, Alfio Gliozzo

Automatically inducing high quality knowledge graphs from a given collection of documents still remains a challenging problem in AI. One way to make headway for this problem is through advancements in a related task known as slot filling. In this task, given an entity query in form of [Entity, Slot, ?], a system is asked to fill the slot by generating or extracting the missing value exploiting evidence extracted from relevant passage(s) in the given document collection. The recent works in the field try to solve this task in an end-to-end fashion using retrieval-based language models. In this paper, we present a novel approach to zero-shot slot filling that extends dense passage retrieval with hard negatives and robust training procedures for retrieval augmented generation models. Our model reports large improvements on both T-REx and zsRE slot filling datasets, improving both passage retrieval and slot value generation, and ranking at the top-1 position in the KILT leaderboard. Moreover, we demonstrate the robustness of our system showing its domain adaptation capability on a new variant of the TACRED dataset for slot filling, through a combination of zero/few-shot learning. We release the source code and pre-trained models.

PaperPDFConference PDFCodeCode Syntology ran

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

No sample from the archive's official repository (ibm/kgi-slot-filling) is linked to this paper; the 8 samples below come from 1 community repository. Syntology harvested code from ibm/kgi-slot-filling for another paper.

ibm/kgi-slot-filling: Syntology's graph links these samples to Re2G: Retrieve, Rerank, Generate (arXiv:2207.06300; that paper's own run record: 7 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; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method.

Syntology 8 code samples harvested from 1 repository linked to this paper: 7 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; 1 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 8 samples from 1 repository, 7 ran (6 with no instrument failure; 1 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 on 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 7 samples that ran has no recorded date.

ibm/kgi-slot-filling officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Re2G: Retrieve, Rerank, Generate (arXiv:2207.06300; that paper's own run record: 7 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; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
IBM/retrieve-write-slot-filling mentioned on GitHubpytorchApache-2.0Syntology: 8 samples, 7 ran (1 instrument-failed) · 1 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

8 samples harvested: 7 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; 1 where Syntology's instrument failed) · 1 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 · our draft was wrong
6ran
1unverified

Licence: 1 of the 8 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, 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 IBM/retrieve-write-slot-filling. “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.

IBM/retrieve-write-slot-filling community (archive-listed) 8 samples, 7 ran (1 instrument-failed) · 1 unverified
clean_text in 84 papers slot_filling/kilt_passage_corpus.py ran fingerprinted Apache-2.0 (permissive) · d890771bd3e4b3f7 · report fixture: signature paragraph:str/0/str · returned str · level recorded 2026-09-25
get_relation_from_inst slot_filling/dataset_stats.py ran Apache-2.0 (permissive) · e57ce1467b8c1988 · report fixture: signature json_inst:dict/0/? · returned str · level recorded 2026-09-25
matching_passage_ids dpr/dpr_kilt_slot_filling_dataset.py ran Apache-2.0 (permissive) · 0fa853d294de8b23 · report fixture: signature pids:list/1/tuple, start_para:int/0/int, end_para:int/0/int · returned list · level recorded 2026-09-25
normalize_answer in 304 papers slot_filling/rag_util.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · e7e75981cb464788 · report fixture: signature s:str/0/str · returned str · level recorded, date not recorded
postprocess_docs slot_filling/rag_util.py ran Apache-2.0 (permissive) · 89afe8d9e930bb10 · report fixture: signature config:MockRagConfig/0/?, tokenizer:MockRagTokenizer/0/?, docs:list/1/dict, input_strings:list/1/str, prefix:str/0/str, n_docs:int/0/int, return_tensors:str/0/str · returned tuple · level recorded 2026-09-25
prefix_allowed_tokens_fn in 6 papers slot_filling/rag_client_server_apply.py ran Apache-2.0 (permissive) · 1d4a7d4d6e3eb7e0 · report fixture: signature gtokenizer:callable/0/?, candidates_text:list/1/str, batch_candidates:list/1/list, batch_id:int/0/int, input_ids:Tensor/1/int · returned list · level recorded 2026-09-25
prepare_seq2seq_batch slot_filling/rag_util.py ran Apache-2.0 (permissive) · f92394bf8b4685cc · report fixture: signature tokenizer:MockTokenizer/0/?, src_texts:list/1/str, tgt_texts:list/1/str, max_length:NoneType/0/?, max_target_length:NoneType/0/?, padding:str/0/str, return_tensors:NoneType/0/?, truncation:bool/0/bool, kwargs:dict/0/? · returned dict · level recorded 2026-09-25
find_para_range slot_filling/kilt_passage_corpus.py unverified Apache-2.0 (permissive) · 50ec835b3753c6ec · report

Tasks

Domain AdaptationFew-Shot LearningKnowledge GraphsPassage RetrievalRetrievalRetrieval-augmented GenerationSlot FillingZero-shot Slot Fillingslot-filling

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Zero-shot Slot Filling T-REx DPRDNS+RAG R-Prec 74.34 #1 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRDNS+RAG R@5 82.89 #1 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRBM25+RAG R-Prec 65.02 #2 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRBM25+RAG R@5 75.52 #2 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRNQ+RAG R-Prec 53.04 #3 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRNQ+RAG R@5 65.54 #3 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRBM25 R-Prec 49.02 #4 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRBM25 R@5 63.34 #4 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRDNS R-Prec 42.62 #5 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRDNS R@5 55.09 #5 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRNQ R-Prec 19.5 #6 of 6 Archive leaderboard report
Zero-shot Slot Filling T-REx DPRNQ R@5 29.8 #6 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRDNS+RAG R-Prec 98.60 #1 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRDNS+RAG R@5 99.70 #1 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRDNS R-Prec 97.53 #2 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRDNS R@5 99.3 #2 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRBM25+RAG R-Prec 96.89 #3 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRBM25+RAG R@5 98.01 #3 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRBM25 R-Prec 94.55 #4 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRBM25 R@5 98.17 #4 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRNQ+RAG R-Prec 68.13 #5 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRNQ+RAG R@5 79.19 #5 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRNQ R-Prec 45.49 #6 of 6 Archive leaderboard report
Zero-shot Slot Filling zsRE DPRNQ R@5 60.77 #6 of 6 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.

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