Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation | Paper Citation Record

Observation ab403ad6-b9c5-4aac-84c0-c3a8660c7342 · outbound

This paper cites A convergence theory for deep learning via over- parameterization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A convergence theory for deep learning via over- parameterization

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.246373Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1fbd1f3-e067-4dde-bbe8-94fde42c6d62 · outbound

This paper cites Amortized im- plicit differentiation for stochastic bilevel optimiza- tion.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Amortized im- plicit differentiation for stochastic bilevel optimiza- tion

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.289759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab1f98d2-f63d-446a-832a-f476dfcdc24d · outbound

This paper cites What Data Enables Optimal Decisions? An Exact Characterization for Linear Optimization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation What Data Enables Optimal Decisions? An Exact Characterization for Linear Optimization

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.341168Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7474ef4b-82da-4094-b24b-17637e419818 · outbound

This paper cites Pythia: A suite for analyzing large language models across training and scaling.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Pythia: A suite for analyzing large language models across training and scaling

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.402085Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 163ffbf2-d36e-4193-98b5-25145af70ccc · outbound

This paper cites Cambridge uni- versity press, 2004.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Cambridge uni- versity press, 2004

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.461175Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18e996b5-4944-4d5c-a3b9-2b7a57a66f66 · outbound

This paper cites Ef- ficient first-order optimization on the pareto set for multi-objective learning under preference guid- ance.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Ef- ficient first-order optimization on the pareto set for multi-objective learning under preference guid- ance

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.514989Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 351dec59-0cc8-4e93-a2c5-9526086b688e · outbound

This paper cites Closing the gap: Tighter analysis of alternating stochastic gradient methods for bilevel problems.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Closing the gap: Tighter analysis of alternating stochastic gradient methods for bilevel problems

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.520878Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3216ebf7-ffd7-4d83-acbb-dc3d61320bb1 · outbound

This paper cites Safety-aware fine-tuning of large language models.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Safety-aware fine-tuning of large language models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.540733Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation de36fc85-c6d9-483c-8fdc-e7d6c92ebcbe · outbound

This paper cites A framework for bilevel op- timization that enables stochastic and global vari- ance reduction algorithms.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A framework for bilevel op- timization that enables stochastic and global vari- ance reduction algorithms

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.639248Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc563609-9254-429f-aec9-94271f718d1f · outbound

This paper cites Solving bilevel multi-objective optimization problems using evolu- tionary algorithms.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Solving bilevel multi-objective optimization problems using evolu- tionary algorithms

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.731383Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a5dab678-5645-4aee-857a-3ee994b18cce · outbound

This paper cites Dempe and P.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Dempe and P

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.777569Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d6c00817-1f34-4c7d-92d0-877129740414 · outbound

This paper cites Springer, 2005.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Springer, 2005

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.897053Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f84fe988-16d4-4402-be4d-8bb37c2abc15 · outbound

This paper cites Mitigating forgetting in llm su- pervised fine-tuning and preference learning.arXiv preprint arXiv:2410.15483, 2024.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Mitigating forgetting in llm su- pervised fine-tuning and preference learning.arXiv preprint arXiv:2410.15483, 2024

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:51.968465Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 76ce55f1-4e95-49b9-9648-2e8209424493 · outbound

This paper cites Forward and reverse gradient-based hyperparameter optimization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Forward and reverse gradient-based hyperparameter optimization

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.073323Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d9f218b1-bde7-4ece-939c-c39b49cadd13 · outbound

This paper cites Bilevel programming for hyperparameter optimization and meta-learning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Bilevel programming for hyperparameter optimization and meta-learning

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.178458Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d17162b0-d216-459e-94a9-f0b8b1f49fbb · outbound

This paper cites Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.214551Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f4509a2b-db27-493f-9081-6cb2077a5473 · outbound

This paper cites On the Properties of the Softmax Function with Application in Game Theory and Reinforcement Learning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation On the Properties of the Softmax Function with Application in Game Theory and Reinforcement Learning

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.231148Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1bd0803c-9180-416c-bbb6-205fec7b80af · outbound

This paper cites The language model evaluation harness, 07 2024.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation The language model evaluation harness, 07 2024

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.313568Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e49ede4-ad42-4151-b71b-3cc9c45626e8 · outbound

This paper cites Learning and data selection in big datasets.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Learning and data selection in big datasets

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.335863Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 052265c8-8825-4c00-94e0-37800b3c3b02 · outbound

This paper cites Approximation Methods for Bilevel Programming.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Approximation Methods for Bilevel Programming

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.337898Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8e2d2b7c-a764-4fa3-b620-67220d4d797a · outbound

This paper cites The Llama 3 Herd of Models.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation The Llama 3 Herd of Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.342653Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 528e2287-f758-4241-b95b-d544da564c91 · outbound

This paper cites On the iteration com- plexity of hypergradient computation.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation On the iteration com- plexity of hypergradient computation

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.344850Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8d7942c8-ebd2-47e6-8635-f1f2cfb3b91f · outbound

This paper cites Reinforced Self-Training (ReST) for Language Modeling.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Reinforced Self-Training (ReST) for Language Modeling

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.346982Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0bef5df0-5185-4d3a-bdf3-a8fd0044c38c · outbound

This paper cites A two-timescale stochastic algo- rithm framework for bilevel optimization: Com- plexity analysis and application to actor-critic.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A two-timescale stochastic algo- rithm framework for bilevel optimization: Com- plexity analysis and application to actor-critic

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.349698Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e96d6a9-1e48-4810-8041-b7091217df71 · outbound

This paper cites User prefer- ence meets pareto-optimality in multi-objective bayesian optimization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation User prefer- ence meets pareto-optimality in multi-objective bayesian optimization

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.351881Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 46ab4710-a689-4bb1-bbd3-3ca176b4bede · outbound

This paper cites Bilevel optimization: Convergence analysis and enhanced design.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Bilevel optimization: Convergence analysis and enhanced design

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.353862Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1f137fbb-69f8-4289-8b7d-c82a6079b679 · outbound

This paper cites Directional conver- gence and alignment in deep learning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Directional conver- gence and alignment in deep learning

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.355984Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 337139c7-ab9c-4b0d-b0f8-71d6626f297e · outbound

This paper cites Get more for less: Principled data selection for warming up fine- tuning in llms.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Get more for less: Principled data selection for warming up fine- tuning in llms

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.358055Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0a5e236b-dffb-41af-b096-d29546ca20be · outbound

This paper cites A near-optimal algorithm for stochastic bilevel op- timization via double-momentum.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A near-optimal algorithm for stochastic bilevel op- timization via double-momentum

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.360446Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fd31897d-b889-4b17-9378-a067f811cb3d · outbound

This paper cites Adam: A method for stochas- tic optimization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Adam: A method for stochas- tic optimization

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.362491Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dbf93589-9b06-4a4b-8a09-95cb87e739e1 · outbound

This paper cites A fully first-order method for stochastic bilevel optimization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A fully first-order method for stochastic bilevel optimization

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.364627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c71d23e5-574f-4876-aa00-06523f30f26b · outbound

This paper cites On penalty methods for non- convex bilevel optimization and first-order stochas- tic approximation.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation On penalty methods for non- convex bilevel optimization and first-order stochas- tic approximation

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.366806Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0527d3e8-655e-4af7-8a7c-0a7a12cef4cc · outbound

This paper cites A fully sin- gle loop algorithm for bilevel optimization without hessian inverse.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A fully sin- gle loop algorithm for bilevel optimization without hessian inverse

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.369033Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 224b8105-1393-4311-b7a7-f48958de1ac8 · outbound

This paper cites Hashimoto.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Hashimoto

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.371120Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb82aa4c-c984-4bae-9043-ca40a3253d32 · outbound

This paper cites Data mixing optimization for supervised fine-tuning of large language models.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Data mixing optimization for supervised fine-tuning of large language models

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.373501Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ef2126c6-d6a9-49b0-8ed7-033f9a685c91 · outbound

This paper cites Preserving diversity in supervised fine-tuning of large language models.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Preserving diversity in supervised fine-tuning of large language models

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.375598Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 85d80849-762c-407c-9ce1-f131abc0ce80 · outbound

This paper cites Data- efficient fine-tuning for llm-based recommenda- tion.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Data- efficient fine-tuning for llm-based recommenda- tion

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.377841Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fb3e154e-0b23-4e74-96b8-ad90e3e23bc4 · outbound

This paper cites Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.379898Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6f674fb5-6687-417d-9068-c957bacb11ec · outbound

This paper cites Regmix: Data mixture as re- gression for language model pre-training.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Regmix: Data mixture as re- gression for language model pre-training

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.382561Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a568a591-d504-450f-b704-2080456295f8 · outbound

This paper cites A generic first-order algo- rithmic framework for bi-level programming be- yond lower-level singleton.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A generic first-order algo- rithmic framework for bi-level programming be- yond lower-level singleton

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.384658Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4f60168-a918-47ec-903e-0e95c2294e06 · outbound

This paper cites Averaged method of multipliers for bi-level optimization without lower-level strong convexity.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Averaged method of multipliers for bi-level optimization without lower-level strong convexity

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.386824Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7a22653e-b4af-415e-9d9c-da8c9e280170 · outbound

This paper cites What makes good data for align- ment? a comprehensive study of automatic data selection in instruction tuning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation What makes good data for align- ment? a comprehensive study of automatic data selection in instruction tuning

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.389047Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7db3ff2-9e74-4d1a-b57b-2a23dafe8a01 · outbound

This paper cites The flan col- lection: Designing data and methods for effective instruction tuning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation The flan col- lection: Designing data and methods for effective instruction tuning

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.391093Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 55c26a91-1207-4c67-95be-2de50e27c814 · outbound

This paper cites First-order penalty methods for bilevel optimization.SIAM Journal on Optimization, 34(2):1937–1969, 2024.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation First-order penalty methods for bilevel optimization.SIAM Journal on Optimization, 34(2):1937–1969, 2024

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.393537Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a0c5299b-dade-46ea-a0fe-33c19989ff8d · outbound

This paper cites Learning what reinforcement learning can’t: Interleaved online fine-tuning for hardest questions.arXiv preprint arXiv:2506.07527, 2025.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Learning what reinforcement learning can’t: Interleaved online fine-tuning for hardest questions.arXiv preprint arXiv:2506.07527, 2025

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.395773Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6b52a6fd-6899-45b4-86f6-87bc822bcb3e · outbound

This paper cites Multi- task learning with user preferences: Gradient de- scent with controlled ascent in pareto optimiza- Offline Data Selection and Online Self-refining Generation tion.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Multi- task learning with user preferences: Gradient de- scent with controlled ascent in pareto optimiza- Offline Data Selection and Online Self-refining Generation tion

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.397840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 417eb84b-5cb5-4243-a5ed-d7f52a923e3d · outbound

This paper cites Emergence of separable manifolds in deep language representations.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Emergence of separable manifolds in deep language representations

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.399939Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b046e0e0-c79e-40d7-8a36-f7c43eefc12d · outbound

This paper cites MIT press, 2018.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation MIT press, 2018

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.402130Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e42ff315-af8b-46e3-bd2f-96526a6a3c38 · outbound

This paper cites Orca: Progressive Learning from Complex Explanation Traces of GPT-4.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Orca: Progressive Learning from Complex Explanation Traces of GPT-4

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.404282Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9fa07a31-73c2-4d30-82c2-6ac92569497c · outbound

This paper cites Token cleaning: Fine-grained data selection for llm su- pervised fine-tuning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Token cleaning: Fine-grained data selection for llm su- pervised fine-tuning

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.406482Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9be8929e-01f9-44a6-8bb8-e7098921251b · outbound

This paper cites Hyperparameter optimization with approximate gradient.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Hyperparameter optimization with approximate gradient

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.408660Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f462f6cf-a878-4e4f-b5f8-3df9e2a174ed · outbound

This paper cites Learning dy- namics of LLM Finetuning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Learning dy- namics of LLM Finetuning

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.410689Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 765a2a02-1b7f-461f-9cfa-194dc25340a1 · outbound

This paper cites Cos-dpo: Condi- tioned one-shot multi-objective fine-tuning frame- work.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Cos-dpo: Condi- tioned one-shot multi-objective fine-tuning frame- work

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.412789Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff9223eb-75e3-42ba-8714-2617c25472f5 · outbound

This paper cites Springer Science & Business Media, 2009.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Springer Science & Business Media, 2009

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.415398Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 989cafcb-8a90-4fd5-9dd1-0f2ba80c5d4d · outbound

This paper cites Large language models encode seman- tics in low-dimensional linear subspaces.arXiv preprint arXiv:2507.09709, 2025.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Large language models encode seman- tics in low-dimensional linear subspaces.arXiv preprint arXiv:2507.09709, 2025

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.418035Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 398ff0eb-37e8-421d-bd04-52221b100a96 · outbound

This paper cites Equilibrium propagation: Bridging the gap between energy- based models and backpropagation.Frontiers in computational neuroscience, 11:24, 2017.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Equilibrium propagation: Bridging the gap between energy- based models and backpropagation.Frontiers in computational neuroscience, 11:24, 2017

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.420099Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 617839a0-e3aa-4645-9e3d-77ea19dd68c8 · outbound

This paper cites Seal: Safety-enhanced aligned llm fine- tuning via bilevel data selection.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Seal: Safety-enhanced aligned llm fine- tuning via bilevel data selection

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.422399Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 76f15c48-928d-4b1b-9ff4-0c060f1750e6 · outbound

This paper cites On penalty-based bilevel gradient descent method.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation On penalty-based bilevel gradient descent method

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.424746Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3df3e21c-8e05-4b89-825b-31ec602607f1 · outbound

This paper cites The importance of online data: Un- derstanding preference fine-tuning via coverage.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation The importance of online data: Un- derstanding preference fine-tuning via coverage

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.426831Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8b2d2a75-0a41-4f78-a641-5f0edbb5bdbf · outbound

This paper cites The im- plicit bias of gradient descent on separable data.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation The im- plicit bias of gradient descent on separable data

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.429041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2f8cf577-fd17-4fa6-95f5-d901e7bb420d · outbound

This paper cites Preference fine-tuning of llms should leverage sub- optimal, on-policy data.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Preference fine-tuning of llms should leverage sub- optimal, on-policy data

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.431056Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 545d82f0-6939-4170-987b-220a7677e883 · outbound

This paper cites New merit functions for multiobjective optimization and their properties.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation New merit functions for multiobjective optimization and their properties

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.433567Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 69fb5c0f-4a33-4c55-9834-14dff3e2c8e7 · outbound

This paper cites Stanford al- paca: An instruction-following llama model, 2023.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Stanford al- paca: An instruction-following llama model, 2023

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.435845Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 93817fb6-e601-4ce0-93cc-2b2fdcd1cfb7 · outbound

This paper cites Transformers as Support Vector Machines.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Transformers as Support Vector Machines

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.437823Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b522613c-7c15-4fff-b3d5-db9978c6b0c5 · outbound

This paper cites Importance sampling: a review.Wiley Interdisciplinary Re- views: Computational Statistics, 2(1):54–60, 2010.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Importance sampling: a review.Wiley Interdisciplinary Re- views: Computational Statistics, 2(1):54–60, 2010

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.440533Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c76cc4eb-7356-443e-b0ec-4c93f20133a8 · outbound

This paper cites Attention is all you need.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Attention is all you need

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.442578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18318a74-416b-478d-82c7-d6d1d623ae1e · outbound

This paper cites On implicit bias in overparameterized bilevel op- timization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation On implicit bias in overparameterized bilevel op- timization

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.444837Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a5a12c8-aefc-401d-aa38-eb2debe966e2 · outbound

This paper cites Interpretable preferences via multi-objective reward modeling and mixture- of-experts.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Interpretable preferences via multi-objective reward modeling and mixture- of-experts

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.446962Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3750258a-053c-4426-8b77-3aa8a4e08d6b · outbound

This paper cites Hbo: Hierarchical balancing op- timization for fine-tuning large language models.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Hbo: Hierarchical balancing op- timization for fine-tuning large language models

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.449234Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c97a6dd-4bc2-4877-ab78-edfd3fa6f655 · outbound

This paper cites Self-Instruct: Aligning Language Models with Self-Generated Instructions.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Self-Instruct: Aligning Language Models with Self-Generated Instructions

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.451438Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bad86346-ed9b-4e2d-9f7d-71230d294e32 · outbound

This paper cites Finetuned language models are zero-shot learners.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Finetuned language models are zero-shot learners

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.454758Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e7def17-8153-4136-8ae6-e2d285534295 · outbound

This paper cites Ldc-mtl: Balancing multi-task learning through scalable loss discrepancy control.arXiv preprint arXiv:2502.08585, 2025.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Ldc-mtl: Balancing multi-task learning through scalable loss discrepancy control.arXiv preprint arXiv:2502.08585, 2025

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.457040Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 022aa620-7218-4883-b983-8c54c2633a33 · outbound

This paper cites Unlocking Global Optimality in Bilevel Optimization: A Pilot Study.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Unlocking Global Optimality in Bilevel Optimization: A Pilot Study

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.459185Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0eac551-d7b8-4950-9efd-31ed4d1df885 · outbound

This paper cites A gen- eralized alternating method for bilevel optimiza- tion under the polyak- lojasiewicz condition.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation A gen- eralized alternating method for bilevel optimiza- tion under the polyak- lojasiewicz condition

Reference 75

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.461439Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7de097fd-993f-41dc-bb6a-3fc3274f7581 · outbound

This paper cites Self-training with noisy student im- proves imagenet classification.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Self-training with noisy student im- proves imagenet classification

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.463444Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 640788d6-774e-4cbe-bde3-f197f2767a47 · outbound

This paper cites Doremi: Optimizing data mixtures speeds up lan- guage model pretraining.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Doremi: Optimizing data mixtures speeds up lan- guage model pretraining

Reference 77

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.466228Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab5008b9-029f-4bf6-901e-d21f65625183 · outbound

This paper cites Data selection for language models via importance resampling.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Data selection for language models via importance resampling

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.468287Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a61cb33b-7a90-42eb-b11b-dfd479d579d5 · outbound

This paper cites Prov- ably faster algorithms for bilevel optimization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Prov- ably faster algorithms for bilevel optimization

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.471064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fea06d46-0954-45c0-8340-9f2ba7c05b2d · outbound

This paper cites LLM Data Selection and Utilization via Dynamic Bi-level Optimization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation LLM Data Selection and Utilization via Dynamic Bi-level Optimization

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.473205Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b663a8ad-5819-4fd6-bd3a-523c052755c7 · outbound

This paper cites Understanding why generalized reweighting does not improve over ERM.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Understanding why generalized reweighting does not improve over ERM

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.475395Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb086deb-3041-4c44-999d-afcca190b44f · outbound

This paper cites Harnessing diversity for important data selection in pretraining large language models.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Harnessing diversity for important data selection in pretraining large language models

Reference 82

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.477689Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c3937c39-b92d-457d-99a2-1bb1a7e65bc6 · outbound

This paper cites Understanding deep learning requires rethinking generalization.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Understanding deep learning requires rethinking generalization

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.479829Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f53ba641-7e99-46e1-a61e-fdf2ced22eb1 · outbound

This paper cites The best instruction-tuning data are those that fit.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation The best instruction-tuning data are those that fit

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.482133Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05c01dc6-f7b7-42a5-872c-3b42d11c03ca · outbound

This paper cites How unlabeled data improve generalization in self-training? a one-hidden-layer theoretical analysis.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation How unlabeled data improve generalization in self-training? a one-hidden-layer theoretical analysis

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.484401Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c2b5febc-8160-4bde-aea1-14ff71e3473f · outbound

This paper cites On-policy rl meets off-policy ex- perts: Harmonizing supervised fine-tuning and re- inforcement learning via dynamic weighting.arXiv preprint arXiv:2508.11408, 2025.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation On-policy rl meets off-policy ex- perts: Harmonizing supervised fine-tuning and re- inforcement learning via dynamic weighting.arXiv preprint arXiv:2508.11408, 2025

Reference 86

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.486696Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9e0bf03-e009-4b44-9754-7c68eefb0a6d · outbound

This paper cites An Introduction to Bi-level Optimization: Foundations and Applications in Signal Processing and Machine Learning.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation An Introduction to Bi-level Optimization: Foundations and Applications in Signal Processing and Machine Learning

Reference 87

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.489056Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3c6b83d0-8f95-4dc5-9580-ab3357ebb7cc · outbound

This paper cites Panacea: Pareto alignment via preference adaptation for llms.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Panacea: Pareto alignment via preference adaptation for llms

Reference 88

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.492479Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ae3612b1-468d-45e7-b1a7-ddc556eea1ff · outbound

This paper cites Lima: Less is more for alignment.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Lima: Less is more for alignment

Reference 89

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.494607Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 414718cb-c6a8-4a23-94af-be5b1a032084 · outbound

This paper cites an unresolved cited work.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Unresolved cited work

Reference 90

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.498005Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4fbf2bf7-48f0-427e-8a50-f7de2fd588ed · outbound

This paper cites an unresolved cited work.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Unresolved cited work

Reference 91

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.500627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 262a2e27-03ee-4b3d-9144-b0e394d2cc0b · outbound

This paper cites an unresolved cited work.

Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation Unresolved cited work

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-03T20:12:52.503592Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers