pith:BNMDKKCE
RAG over Thinking Traces Can Improve Reasoning Tasks
Retrieving thinking traces from problem-solving attempts improves reasoning performance on math and code benchmarks.
arxiv:2605.03344 v2 · 2026-05-05 · cs.IR · cs.AI · cs.CL
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{BNMDKKCE6ACDCGB24WV6QR2UHK}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
Using these traces as a corpus, a simple retrieve-then-generate pipeline consistently improves reasoning performance across strong models and benchmarks such as AIME 2025--2026, LiveCodeBench, and GPQA-Diamond, outperforming both non-RAG baselines and retrieval over standard web corpora.
That thinking traces generated during problem-solving attempts contain generalizable, high-quality reasoning signals that transfer usefully to new problems and different models without introducing systematic errors or biases from the trace-generation process itself.
RAG over structured thinking traces boosts LLM reasoning on AIME, LiveCodeBench, and GPQA, with relative gains up to 56% and little added cost.
Cited by
Receipt and verification
| First computed | 2026-06-10T01:08:36.133373Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
0b58352844f00431183ae5abe847543aa8904888c20febfc08d92574fc77f68a
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/BNMDKKCE6ACDCGB24WV6QR2UHK \
| 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: 0b58352844f00431183ae5abe847543aa8904888c20febfc08d92574fc77f68a
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "bfbff81563853940b3c3a52b262a2bcd53b55bdf33a5b2c13ec22862c7c421c8",
"cross_cats_sorted": [
"cs.AI",
"cs.CL"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.IR",
"submitted_at": "2026-05-05T04:03:28Z",
"title_canon_sha256": "5e890a4c40583b848108dcf231ec24392004f1045164a61c171a89be9de09ef2"
},
"schema_version": "1.0",
"source": {
"id": "2605.03344",
"kind": "arxiv",
"version": 2
}
}