Pith. sign in

REVIEW 10 cited by

Self-Alignment with Instruction Backtranslation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2308.06259 v3 pith:VJNDN3E4 submitted 2023-08-11 cs.CL

classification cs.CL
keywords modelinstructiondataapproachbacktranslationexampleshighlanguage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present a scalable method to build a high quality instruction following language model by automatically labelling human-written text with corresponding instructions. Our approach, named instruction backtranslation, starts with a language model finetuned on a small amount of seed data, and a given web corpus. The seed model is used to construct training examples by generating instruction prompts for web documents (self-augmentation), and then selecting high quality examples from among these candidates (self-curation). This data is then used to finetune a stronger model. Finetuning LLaMa on two iterations of our approach yields a model that outperforms all other LLaMa-based models on the Alpaca leaderboard not relying on distillation data, demonstrating highly effective self-alignment.

Discussion (0). Sign in to comment.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. HSS-Synth: Humanities and Social Sciences Data Synthesis for LLMs

    cs.CL 2026-07 conditional novelty 6.0 of 10

    HSS-Synth generates 230k instruction-tuning samples for 14 humanities/social-science fields and reports state-of-the-art fine-tuning results on 16 benchmarks.

  2. Bridging Offline and Online Reinforcement Learning for LLMs

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Semi-online DPO, which syncs the generation model every few update steps, performs nearly as well as fully online DPO and GRPO, while strongly beating offline DPO.

  3. Aligning Large Language Models with Implicit Preferences from User-Generated Content

    cs.CL 2025-06 conditional novelty 6.0 of 10

    PUGC turns unlabeled user-generated content into preference pairs by scoring policy responses against the source post as a reference, reaching 35.93% length-controlled win rate on AlpacaEval 2.0 with Mistral-7B-Instruct.

  4. From Real to Synthetic: Synthesizing Millions of Diversified and Complicated User Instructions with Attributed Grounding

    cs.CL 2025-06 conditional novelty 6.0 of 10

    A 1M-instruction dataset, generated by attributing real instructions to documents, users and motivations and then reversing the process over web documents, improves instruction-following benchmarks for 7B-14B models.

  5. Can Structured Templates Facilitate LLMs in Tackling Harder Tasks? : An Exploration of Scaling Laws by Difficulty

    cs.AI 2025-08 reject novelty 5.0 of 10

    Training on easy synthetic math data lowers accuracy on hard benchmarks, and the proposed SST framework, which teaches explicit procedural chains, aims to reverse that drop.

  6. Temporal Self-Rewarding Language Models: Decoupling Chosen-Rejected via Past-Future

    cs.CL 2025-08 conditional novelty 5.0 of 10

    Anchoring rejected responses to the initial model and choosing responses from a future model raises AlpacaEval 2.0 win rate from 19.69 to 29.44 for Llama3.1-8B.

  7. An Uncertainty-Driven Adaptive Self-Alignment Framework for Large Language Models

    cs.AI 2025-07 conditional novelty 5.0 of 10

    A training pipeline that scores a model's own responses for semantic, factual, and safety uncertainty, builds preference pairs from those scores, and trains in three difficulty stages improves reported alignment score...

  8. ALPS: Attention Localization and Pruning Strategy for Efficient Alignment of Large Language Models

    cs.CL 2025-05 conditional novelty 5.0 of 10

    ALPS selects task-sensitive attention heads by measuring Wasserstein distance between base and task-tuned weights, and freezing other heads during fine-tuning improves performance and efficiency.

  9. Mixtures of SubExperts for Large Language Continual Learning

    cs.LG 2025-11 reject novelty 4.0 of 10

    MoSEs combines task-specific prompts, keys, and sparse low-rank subexpert masks with a router to cut catastrophic forgetting in sequential LLM fine-tuning.

  10. Can Compressed LLMs Truly Act? An Empirical Evaluation of Agentic Capabilities in LLM Compression

    cs.LG 2025-05 conditional novelty 4.0 of 10

    ACBench tests compressed LLMs on agentic tasks and finds 4-bit quantization keeps tool use and workflow generation strong while hurting real-world application performance.

Pith tools