newsHacker NewsTrust 52 · CommunityPublished 1mo agoLive · 1mo ago

Knowledge Distillation of Black-Box Large Language Models

48points12comments

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

LinkedLinked via unknownHow Surprising Is Historical Italian to Language Models? Tokenization Tax, Comprehension Tax, and a Simple Mitigation →
LinkedLinked via unknownNuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models →
LinkedLinked via unknownTransformer →
LinkedLinked via unknownengineering87/llm-atlas →
LinkedLinked via unknownminimal-diffusion-lm →
LinkedLinked via unknownWhen are likely answers right? On Sequence Probability and Correctness in LLMs →
LinkedLinked via unknownScaling limit of the Random Language Model →
LinkedLinked via unknownAB-RAG: Adaptive Budgeted Retrieval-Augmented Generation for Reliable Question Answering →

Covers

paperHow Surprising Is Historical Italian to Language Models? Tokenization Tax, Comprehension Tax, and a Simple Mitigation paperNuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models glossary_termTransformer repoengineering87/llm-atlas repominimal-diffusion-lm

Covers (incoming)

paperWhen are likely answers right? On Sequence Probability and Correctness in LLMs paperScaling limit of the Random Language Model paperAB-RAG: Adaptive Budgeted Retrieval-Augmented Generation for Reliable Question Answering paperLittle Brains, Big Feats: Exploring Compact Language Models paperEfficient Retrieval-Augmented Generation via Token Co-occurrence Graphs paperGrounding LLM Reasoning under Incomplete Graph Evidence paperThe Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training Methodology repochrisliu298/awesome-on-policy-distillation repochrisliu298/awesome-llm-unlearning paperObject Aligner: A Configurable JSON Schema Similarity Score for Graphs, Applied to LLM Prompt Optimization paperEduArt: An educational-level benchmark for evaluating art history knowledge in large language models reponick7nlp/Awesome-LLM-On-Policy-Distillation paperThe One-Word Census: Answer-Choice Conformity Across 44 Language Models paperDebate-on-Graph: Reliable and Adaptive Reasoning of Large Language Model on Uncertain Knowledge Graph papersurprisal is Not a Theory reposileod/llm-theory-of-mind paperLanguage Identification with Succinct Machine-Independent Traces repotyang816/Awesome-TCM-LLM repop-e-w/heretic repollmsresearch/llm-flashcards paperGlobally Consistent Coloring Schemes for Language Identification paperHow Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation?

Related across the graph

repochrisliu298/awesome-llm-unlearning paperDebate-on-Graph: Reliable and Adaptive Reasoning of Large Language Model on Uncertain Knowledge Graph paperLittle Brains, Big Feats: Exploring Compact Language Models paperObject Aligner: A Configurable JSON Schema Similarity Score for Graphs, Applied to LLM Prompt Optimization repochrisliu298/awesome-on-policy-distillation paperWhen are likely answers right? On Sequence Probability and Correctness in LLMs paperLanguage Identification with Succinct Machine-Independent Traces glossary_termTransformer repominimal-diffusion-lm repoengineering87/llm-atlas paperAB-RAG: Adaptive Budgeted Retrieval-Augmented Generation for Reliable Question Answering paperGrounding LLM Reasoning under Incomplete Graph Evidence reponick7nlp/Awesome-LLM-On-Policy-Distillation repotyang816/Awesome-TCM-LLM paperGlobally Consistent Coloring Schemes for Language Identification repollmsresearch/llm-flashcards repop-e-w/heretic paperEfficient Retrieval-Augmented Generation via Token Co-occurrence Graphs paperThe One-Word Census: Answer-Choice Conformity Across 44 Language Models paperEduArt: An educational-level benchmark for evaluating art history knowledge in large language models paperHow Surprising Is Historical Italian to Language Models? Tokenization Tax, Comprehension Tax, and a Simple Mitigation paperHow Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation?papersurprisal is Not a Theory paperThe Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training Methodology paperNuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models paperScaling limit of the Random Language Model reposileod/llm-theory-of-mind