Pith Number

pith:JDLC77MY

pith:2026:JDLC77MYMM6ZCECZ6Z52HEZ2FX

not attested not anchored not stored refs pending

Federation over Text: Insight Sharing for Multi-Agent Reasoning

Dixi Yao, Manzil Zaheer, Tahseen Rabbani, Tian Li

LLM agents can share distilled reasoning traces across tasks to build a reusable library of metacognitive insights that raises accuracy and cuts token use.

arxiv:2604.16778 v2 · 2026-04-18 · cs.LG · cs.AI

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{JDLC77MYMM6ZCECZ6Z52HEZ2FX}

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

Experiments show that FoT improves reasoning effectiveness and efficiency across a wide range of challenging applications... Specifically, it improves average accuracies of downstream tasks by 24% while reducing the reasoning tokens by 28% across the first two applications. In the research insight discovery application, FoT is able to generate insights that cover over 90% of the major contributions in the subsequent papers.

C2weakest assumption

That a central server can reliably aggregate and distill raw reasoning traces from heterogeneous tasks into a compact, cross-domain library of metacognitive insights that actually improve downstream agent performance without any supervision signal or gradient information.

C3one line summary

FoT lets multiple LLM agents federate text-based reasoning traces into a cross-task insight library, raising average task accuracy by 24% and cutting reasoning tokens by 28%.

Receipt and verification

First computed	2026-05-26T01:03:30.212953Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

48d62ffd98633d911059f67ba3933a2ddf0aaaff845b1368697cd2723e50849a

Aliases

arxiv: 2604.16778 · arxiv_version: 2604.16778v2 · doi: 10.48550/arxiv.2604.16778 · pith_short_12: JDLC77MYMM6Z · pith_short_16: JDLC77MYMM6ZCECZ · pith_short_8: JDLC77MY

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/JDLC77MYMM6ZCECZ6Z52HEZ2FX \
  | 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: 48d62ffd98633d911059f67ba3933a2ddf0aaaff845b1368697cd2723e50849a

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "9275a356f7165abd601e0f9a09a33561c7ea49421cb89e5440c12b6017fb3de4",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-04-18T01:57:42Z",
    "title_canon_sha256": "258ff9d8e5866f4d8845297de6e31d47571f27bcb3167138200315e88622598c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.16778",
    "kind": "arxiv",
    "version": 2
  }
}