SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.16705 · NeurIPS · 2024

SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-Reflection

Xuebo Liu, Derek Wong, Ziyi Wang, Min Zhang, Liangxin Liu, Dongfang Li, Baotian Hu

arXiv · PDF · Open in the Atlas

Code that ran

14 of 16 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 13 failures of our instrument; 2 no recorded run), measured 2026-11-21; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2402.16705.json · badge: /api/ran/2402.16705.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Blue-Raincoat/SelectIT canonical (classifier label)
classifier basis: census
12 ran · 2 no recorded run
blue-raincoat/selectit canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
construction_rps Ran
ran; our contract draft was wrong, not the code
blue-raincoat/selectit/self_reflection/token_level.py
pointer only (licence: NONE) · get_code("ccc3a0c10e2aab36")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, blue-raincoat/selectit when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

construction_rps Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/self_reflection/model_level.py
pointer only (licence: NONE) · get_code("977b7a1df6a88a60")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

create_prompt_with_llama2_chat_format Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/eval/templates.py
pointer only (licence: NONE) · get_code("5637030b6f66a7c5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

create_prompt_with_olmo_chat_format Ran
ran on a synthesized input
Blue-Raincoat/SelectIT/eval/eval/templates.py
pointer only (licence: NONE); the stored copy get_code serves, from THU-KEG/ADELIE, is pointer only (licence: NONE) · get_code("45e2cc1022c95249")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

create_prompt_with_tulu_chat_format Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/eval/templates.py
pointer only (licence: NONE) · get_code("661a1d8a5c30c318")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

dispatch_openai_chat_requesets Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/eval/dispatch_openai_requests.py
pointer only (licence: NONE); the stored copy get_code serves, from thu-keg/crab, is pointer only (licence: NONE) · get_code("dcad9ac62979ca49")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

dispatch_openai_prompt_requesets Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/eval/dispatch_openai_requests.py
pointer only (licence: NONE); the stored copy get_code serves, from thu-keg/crab, is pointer only (licence: NONE) · get_code("71e4259026911444")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

encode_with_messages_format Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/open_instruct/finetune.py
pointer only (licence: NONE) · get_code("97229c97d4273595")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

encode_with_messages_format Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/open_instruct/dpo_tune.py
pointer only (licence: NONE) · get_code("254ab4972fe2c4e2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

generate_completions Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Blue-Raincoat/SelectIT/eval/eval/utils.py
pointer only (licence: NONE) · get_code("cdbcc01805fac313")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_acceptance_results Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/human_eval/compute_metrics.py
pointer only (licence: NONE); the stored copy get_code serves, from yikee/FLIP, is pointer only (licence: NONE) · get_code("3db002d1cc62db4b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_comparison_results Ran
ran; our contract draft was wrong, not the code
Blue-Raincoat/SelectIT/eval/human_eval/compute_metrics.py
pointer only (licence: NONE); the stored copy get_code serves, from yikee/FLIP, is pointer only (licence: NONE) · get_code("8eb7a7a7cfdb631c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_next_word_predictions Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Blue-Raincoat/SelectIT/eval/eval/utils.py
pointer only (licence: NONE) · get_code("d134c88af0236263")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

rating Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
blue-raincoat/selectit/self_reflection/model_level.py
pointer only (licence: NONE) · get_code("5d450db4741b3ef3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, blue-raincoat/selectit when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

encode_with_prompt_completion_format Not yet run
unverified (harvested, no recorded run)
Blue-Raincoat/SelectIT/eval/open_instruct/finetune.py
pointer only (licence: NONE) · get_code("119cad9954acc960")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

score_completions Not yet run
unverified (harvested, no recorded run)
Blue-Raincoat/SelectIT/eval/eval/utils.py
pointer only (licence: NONE); the stored copy get_code serves, from jinzhuoran/rwku, is pointer only (licence: NONE) · get_code("d029fa02758da921")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Blue-Raincoat/SelectIT when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 13 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 13 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_next_word_predictions 10 other papers
2 dated earlier, 8 later
2401.15006 (2024-01)
Airavata: Introducing Hindi Instruction-tuned LLM
2401.16405 (2024-01)
Scaling Sparse Fine-Tuning to Large Language Models
2405.05008 (2024-05)
ADELIE: Aligning Large Language Models on Information Extraction
and 7 more not named here
dispatch_openai_prompt_requesets 8 other papers
4 dated earlier, 4 later
2310.06830 (2023-10)
Lemur: Harmonizing Natural Language and Code for Language Agents
2401.10768 (2024-01)
Knowledge Verification to Nip Hallucination in the Bud
2401.15006 (2024-01)
Airavata: Introducing Hindi Instruction-tuned LLM
and 5 more not named here
dispatch_openai_chat_requesets 7 other papers
4 dated earlier, 3 later
2310.06830 (2023-10)
Lemur: Harmonizing Natural Language and Code for Language Agents
2401.10768 (2024-01)
Knowledge Verification to Nip Hallucination in the Bud
2401.15006 (2024-01)
Airavata: Introducing Hindi Instruction-tuned LLM
and 4 more not named here
score_completions 6 other papers
6 later
2405.05008 (2024-05)
ADELIE: Aligning Large Language Models on Information Extraction
2406.10890 (2024-06)
RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models
2410.24175 (2024-10)
Constraint Back-translation Improves Complex Instruction Following of Large Language Models
and 3 more not named here
generate_completions 5 other papers
5 later
2405.05008 (2024-05)
ADELIE: Aligning Large Language Models on Information Extraction
2410.24175 (2024-10)
Constraint Back-translation Improves Complex Instruction Following of Large Language Models
2504.11393 (2025-04)
How to Predict Best Pretraining Data with Small Experiments
and 2 more not named here
encode_with_prompt_completion_format 4 other papers
4 later
2405.05008 (2024-05)
ADELIE: Aligning Large Language Models on Information Extraction
2405.20974 (2024-05)
SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationales
2406.15444 (2024-06)
Published at ICLR 2025 Workshop on Reasoning and Planning for LLMs CUTTING THROUGH THE NOISE: BOOSTING LLM PERFORMANCE ON MATH WORD PROBLEMS
and 1 more not named here
encode_with_messages_format 3 other papers
3 later
2405.05008 (2024-05)
ADELIE: Aligning Large Language Models on Information Extraction
2406.15444 (2024-06)
Published at ICLR 2025 Workshop on Reasoning and Planning for LLMs CUTTING THROUGH THE NOISE: BOOSTING LLM PERFORMANCE ON MATH WORD PROBLEMS
2410.16208 (2024-10)
Compute-Constrained Data Selection
create_prompt_with_llama2_chat_format 2 other papers
2 later
2405.05008 (2024-05)
ADELIE: Aligning Large Language Models on Information Extraction
2505.07437 (2025-05)
LEAD: Iterative Data Selection for Efficient LLM Instruction Tuning
create_prompt_with_olmo_chat_format 2 other papers
2 later
2405.05008 (2024-05)
ADELIE: Aligning Large Language Models on Information Extraction
2505.07437 (2025-05)
LEAD: Iterative Data Selection for Efficient LLM Instruction Tuning
create_prompt_with_tulu_chat_format 1 other paper
1 later
2505.07437 (2025-05)
LEAD: Iterative Data Selection for Efficient LLM Instruction Tuning
encode_with_messages_format 1 other paper
1 later
2410.16208 (2024-10)
Compute-Constrained Data Selection
get_acceptance_results 1 other paper
1 later
2602.13551 (2026-02)
Small Reward Models via Backward Inference
get_comparison_results 1 other paper
1 later
2602.13551 (2026-02)
Small Reward Models via Backward Inference

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Instruction tuning (IT) is crucial to tailoring large language models (LLMs) towards human-centric interactions. Recent advancements have shown that the careful selection of a small, high-quality subset of IT data can significantly enhance the performance of LLMs. Despite this, common approaches often rely on additional models or data, which increases costs and limits widespread adoption. In this work, we propose a novel approach, termed SelectIT, that capitalizes on the foundational capabilities of the LLM itself. Specifically, we exploit the intrinsic uncertainty present in LLMs to more effectively select high-quality IT data, without the need for extra resources. Furthermore, we introduce a curated IT dataset, the Selective Alpaca, created by applying SelectIT to the Alpaca-GPT4 dataset. Empirical results demonstrate that IT using Selective Alpaca leads to substantial model ability enhancement. The robustness of SelectIT has also been corroborated in various foundation models and domain-specific tasks. Our findings suggest that longer and more computationally intensive IT data may serve as superior sources of IT, offering valuable insights for future research in this area. Data, code, and scripts are freely available at https://github.com/Blue-Raincoat/SelectIT.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2402.16705")
get_code_for_paper("2402.16705")
have("2402.16705")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2402.16705.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2402.16705.svg)](https://syntology.ai/paper/2402.16705)

Connect an agent — have() is free.