Pith. sign in
Pith Number

pith:S2TJTMPP

pith:2023:S2TJTMPPYHW32D5ESW4VJ672LC
not attested not anchored not stored refs resolved

Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection

Akari Asai, Avirup Sil, Hannaneh Hajishirzi, Yizhong Wang, Zeqiu Wu

Self-RAG trains a single language model to adaptively retrieve passages on demand and critique its own outputs using special reflection tokens.

arxiv:2310.11511 v1 · 2023-10-17 · cs.CL · cs.AI · cs.LG

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{S2TJTMPPYHW32D5ESW4VJ672LC}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp
2 Internet Archive
3 Author claim open · sign in to claim
4 Citations open
5 Replications open
Portable graph bundle live · download bundle · merged state
The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Self-RAG (7B and 13B parameters) significantly outperforms state-of-the-art LLMs and retrieval-augmented models on a diverse set of tasks. Specifically, Self-RAG outperforms ChatGPT and retrieval-augmented Llama2-chat on Open-domain QA, reasoning and fact verification tasks, and it shows significant gains in improving factuality and citation accuracy for long-form generations relative to these models.

C2weakest assumption

That an arbitrary LM can be trained to generate and act on reflection tokens in a way that improves rather than degrades performance across tasks without introducing new failure modes from the added tokens.

C3one line summary

Self-RAG trains LLMs to adaptively retrieve passages on demand and self-critique using reflection tokens, outperforming ChatGPT and retrieval-augmented Llama2 on QA, reasoning, and fact verification.

References

154 extracted · 154 resolved · 23 Pith anchors

[1] Learning to retrieve reasoning paths over wikipedia graph for question answering 2020
[2] Retrieval-based language models and applications 2023
[3] Task-aware retrieval with instructions 2023
[7] Flashattention: Fast and memory-efficient exact attention with io-awareness 2022
[8] Chain-of-Verification Reduces Hallucination in Large Language Models 2023 · arXiv:2309.11495

Formal links

3 machine-checked theorem links

Cited by

81 papers in Pith

Receipt and verification
First computed 2026-07-05T07:02:08.374924Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

96a699b1efc1edbd0fa495b954fbfa58bc4baefed86c2e351038dcb7ef6897de

Aliases

arxiv: 2310.11511 · arxiv_version: 2310.11511v1 · doi: 10.48550/arxiv.2310.11511 · pith_short_12: S2TJTMPPYHW3 · pith_short_16: S2TJTMPPYHW32D5E · pith_short_8: S2TJTMPP
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/S2TJTMPPYHW32D5ESW4VJ672LC \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 96a699b1efc1edbd0fa495b954fbfa58bc4baefed86c2e351038dcb7ef6897de
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "561e44b2432b4f4e957118a069dde7fae8fc9bfbf8aac408de3d4dfd1beed17d",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.LG"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2023-10-17T18:18:32Z",
    "title_canon_sha256": "4a2dd2ad87d2a6160674ba66eff6276562daa1a531df578df1f5f1e675a0628c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2310.11511",
    "kind": "arxiv",
    "version": 1
  }
}