Papers › GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extraction

GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extraction

5 Oct 2023arXiv:2310.03668archive 2025-07-28

Oscar Sainz, Iker García-Ferrero, Rodrigo Agerri, Oier Lopez de Lacalle, German Rigau, Eneko Agirre

Large Language Models (LLMs) combined with instruction tuning have made significant progress when generalizing to unseen tasks. However, they have been less successful in Information Extraction (IE), lagging behind task-specific models. Typically, IE tasks are characterized by complex annotation guidelines that describe the task and give examples to humans. Previous attempts to leverage such information have failed, even with the largest models, as they are not able to follow the guidelines out of the box. In this paper, we propose GoLLIE (Guideline-following Large Language Model for IE), a model able to improve zero-shot results on unseen IE tasks by virtue of being fine-tuned to comply with annotation guidelines. Comprehensive evaluation empirically demonstrates that GoLLIE is able to generalize to and follow unseen guidelines, outperforming previous attempts at zero-shot information extraction. The ablation study shows that detailed guidelines are key for good results.

PaperPDFCodeCode 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 1 repository linked to this paper: 18 ran (of which 2 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 1 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 25 samples from 1 repository, 18 ran (17 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.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

hitz-zentroa/gollie officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 25 samples, 18 ran (1 instrument-failed) · 7 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.

Code Syntology ran Syntology

25 samples harvested: 18 ran (of which 2 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 1 where Syntology's instrument failed) · 7 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
17ran
7unverified

Licence: 0 of the 25 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from hitz-zentroa/gollie. “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”. For 1 sample that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

hitz-zentroa/gollie official repository 25 samples, 18 ran (1 instrument-failed) · 7 unverified
FlashRotaryEmbedding src/model/patch_models/modeling_flash_llama.py ran Apache-2.0 (permissive) · c8bee51bf997140e · report fixture: returned FlashRotaryEmbedding · level recorded 2026-09-22
LlamaRMSNorm in 9 papers src/model/patch_models/modeling_flash_llama.py ran Apache-2.0 (permissive) · 3e5e1a1a6ae02184 · report fixture: returned LlamaRMSNorm · level recorded 2026-09-22
clean_guidelines src/paraphrase/utils.py ran Apache-2.0 (permissive) · a31cb9b49aa3fb29 · report fixture: signature guidelines:dict/0/? · returned dict · level recorded 2026-09-24
find_all_linear_names in 50 papers src/model/model_utils.py ran Apache-2.0 (permissive) · ef50b7a39a5c925f · report fixture: signature model:callable/0/?, quantization:NoneType/0/? · returned list · level recorded 2026-09-24
fix_prompt_outputs src/evaluate.py ran fingerprinted Apache-2.0 (permissive) · 25ec67d8318f45ea · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-24
get_class in 12 papers src/evaluate.py ran Apache-2.0 (permissive) · 40c63137e2539113 · report fixture: signature class_path:str/0/str · returned type · level recorded 2026-09-24
get_class in 12 papers src/generate_data.py ran Apache-2.0 (permissive) · d5e035b43805bee4 · report fixture: signature class_path:str/0/str · returned type · level recorded 2026-09-24
get_device_map src/model/load_model.py ran Apache-2.0 (permissive) · 47660e151147cb9e · report fixture: signature force_auto_device_map:bool/0/bool, max_memory_MB:int/0/int, use_better_transformer:bool/0/bool · returned tuple · level recorded 2026-09-24
get_num_return_sentences src/paraphrase/utils.py ran Apache-2.0 (permissive) · d4149ebae2786013 · report fixture: signature config_path:str/0/str · returned int · level recorded 2026-09-24
get_trainable_parameters src/model/model_utils.py ran Apache-2.0 (permissive) · f32e00b404c9eec4 · report fixture: signature model:callable/0/? · returned tuple · level recorded 2026-09-24
prepare_data in 87 papers src/dataset/dataset.py ran Apache-2.0 (permissive) · 67c3b2ef9e5c371c · report fixture: signature example:str/0/str, tokenizer:callable/0/?, is_encoder_decoder:bool/0/bool, max_length:int/0/int, inference:bool/0/bool, prompt_loss_weight:float/0/float · returned BatchEncoding · level recorded 2026-09-24
prepare_model_for_kbit_training in 5 papers src/model/model_utils.py ran Apache-2.0 (permissive) · 4fa8a9b51ecfca0c · report fixture: signature model:callable/0/?, use_gradient_checkpointing:bool/0/bool, layer_norm_names:list/1/str, output_embedding_layer_name:str/0/str · returned MockModel · level recorded 2026-09-24
remove_hallucinations src/evaluate.py ran Apache-2.0 (permissive) · 02c4c2306e3a0610 · report fixture: signature unlabelled_sentence:str/0/str, predictions:list/1/Prediction · returned list · level recorded 2026-09-24
rmsnorm_func in 9 papers src/model/patch_models/modeling_flash_llama.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 349dce8e37362617 · report fixture: signature hidden_states:Tensor/2/float, weight:Tensor/1/float, variance_epsilon:float/0/float · returned Tensor · level recorded 2026-09-22
to_bilou_encoding src/tasks/label_encoding.py ran Apache-2.0 (permissive) · d2660d11a380300e · report fixture: signature tags:list/1/str · returned list · level recorded 2026-09-24
to_iob2_encoding src/tasks/label_encoding.py ran Apache-2.0 (permissive) · 6c7118a2f5cc50e4 · report fixture: signature tags:list/1/str · returned list · level recorded 2026-09-24
to_iob_encoding src/tasks/label_encoding.py ran Apache-2.0 (permissive) · f709c942540a746b · report fixture: signature tags:list/1/str · returned list · level recorded 2026-09-24
update_guidelines src/paraphrase/utils.py ran Apache-2.0 (permissive) · e05fd99ba3c452ba · report fixture: signature paraphrases:list/1/str, guidelines:dict/0/?, language:str/0/str · returned dict · level recorded 2026-09-24
LlamaAttention in 15 papers src/model/patch_models/modeling_flash_llama.py unverified Apache-2.0 (permissive) · 04267dd6afc2cfd2 · report
LlamaDecoderLayer in 11 papers src/model/patch_models/modeling_flash_llama.py unverified Apache-2.0 (permissive) · 71e315c10d6d705e · report
LlamaForCausalLM src/model/patch_models/modeling_flash_llama.py unverified Apache-2.0 (permissive) · fb6a17240fa2036c · report
LlamaModel in 9 papers src/model/patch_models/modeling_flash_llama.py unverified Apache-2.0 (permissive) · b6c0bf260825919c · report
batch_tokenization src/dataset/dataset.py unverified Apache-2.0 (permissive) · d28dcf85671df49b · report
load_model in 479 papers src/model/load_model.py unverified Apache-2.0 (permissive) · bfca1181a79266f6 · 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.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for GoLLIE (graph method: GoLLIE). 9 bricks; individually on a fixture: 4 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; 1 where Syntology's instrument failed) · 5 unverified. The assembly has not been run end-to-end.

  1. LlamaForCausalLM entry · level 0 · unverified · fb6a17240fa2036c
  2. LlamaModel model · level 0 · unverified · b6c0bf260825919c
  3. LlamaDecoderLayer model · level 0 · unverified · 71e315c10d6d705e
  4. LlamaRMSNorm model · level 1 · ran · 3e5e1a1a6ae02184
  5. LlamaAttention model · level 0 · unverified · 04267dd6afc2cfd2
  6. LlamaMLP model · level 1 · ran · c5d348f0b0059642
  7. rmsnorm_func helper · level 1 · ran · our draft was wrong · 349dce8e37362617
  8. FlashRotaryEmbedding model · level 1 · ran · c8bee51bf997140e
  9. repeat_kv helper · level 0 · unverified · 44b3ce4579f7806b

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="2310.03668") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Event Argument ExtractionEvent ExtractionLanguage ModelingLanguage ModellingLarge Language ModelLow Resource Named Entity RecognitionNamed Entity RecognitionNamed Entity Recognition (NER)Relation ExtractionZero-shot Event ExtractionZero-shot Named Entity Recognition (NER)

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Named Entity Recognition (NER) ACE 2005 GoLLIE F1 89.6 #2 of 20 Archive leaderboard report
Named Entity Recognition (NER) BC5CDR GoLLIE F1 88.4 #12 of 16 Archive leaderboard report
Named Entity Recognition (NER) CoNLL 2003 (English) GoLLIE F1 93.1 #29 of 73 Archive leaderboard report
Named Entity Recognition (NER) NCBI-disease GoLLIE F1 86.5 #22 of 26 Archive leaderboard report
Named Entity Recognition (NER) WNUT 2017 GoLLIE F1 54.3 #11 of 23 Archive leaderboard report
Relation Extraction ACE 2005 GoLLIE RE Micro F1 70.1 #4 of 30 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) Broad Twitter Corpus GoLLIE Entity F1 51.4 #2 of 2 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) CrossNER GoLLIE AI 61.6 #2 of 4 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) CrossNER GoLLIE Literature 62.7 #2 of 4 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) CrossNER GoLLIE Music 68.4 #2 of 4 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) CrossNER GoLLIE Politics 60.2 #2 of 4 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) CrossNER GoLLIE Science 56.3 #2 of 4 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) HarveyNER GoLLIE Entity F1 41.3 #1 of 2 Archive leaderboard report
Zero-shot Named Entity Recognition (NER) WikiEvents GoLLIE Entity F1 81.3 #1 of 2 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