Pith Number

pith:KG4CXV2S

pith:2026:KG4CXV2S5DJ6NJPIEXYOBBCU33

not attested not anchored not stored refs resolved

Pramana: Fine-Tuning Large Language Models for Epistemic Reasoning through Navya-Nyaya

Sharath Sathish

Fine-tuning language models on Navya-Nyaya logic produces 100% semantic correctness in reasoning even when output format is only partly followed.

arxiv:2604.04937 v1 · 2026-02-14 · cs.AI · cs.CL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{KG4CXV2S5DJ6NJPIEXYOBBCU33}

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

Stage 1 achieves 100% semantic correctness on held-out evaluation despite only 40% strict format adherence revealing that models internalize reasoning content even when structural enforcement is imperfect.

C2weakest assumption

That fine-tuning on only 55 Navya-Nyaya-structured logical problems will instill generalizable epistemic reasoning skills that outperform standard prompting or fine-tuning methods on broader tasks.

C3one line summary

Fine-tuning LLMs on Navya-Nyaya's six-phase reasoning structure yields 100% semantic correctness on held-out logical problems despite only 40% strict format adherence.

References

19 extracted · 19 resolved · 5 Pith anchors

[1] GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models · arXiv:2410.05229

[2] doi: 10.1007/s10781-020-09419-0. Z. Chen et al. Proof of thought: Neurosymbolic program synthesis allows robust and interpretable reasoning. arXiv preprint arXiv:2409.17270, · doi:10.1007/s10781-020-09419-0

[3] Leonardo de Moura and Nikolaj Bjørner 2020 · doi:10.24963/ijcai.2020/538

[4] In: TACAS ’08 · doi:10.1007/978-3-540-78800-3

[5] Revisiting group relative policy op- timization: Insights into on-policy and off-policy training.arXiv preprint arXiv:2505.22257

Formal links

2 machine-checked theorem links

Cited by

1 paper in Pith

Philosophical Dispositions as Behavioral Constraints for AI-Assisted Code Review: An Empirical Study

Receipt and verification

First computed	2026-06-02T02:04:17.143309Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

51b82bd752e8d3e6a5e825f0e08454dee89c6d78b54487bb38d8f51d6c89975f

Aliases

arxiv: 2604.04937 · arxiv_version: 2604.04937v1 · doi: 10.48550/arxiv.2604.04937 · pith_short_12: KG4CXV2S5DJ6 · pith_short_16: KG4CXV2S5DJ6NJPI · pith_short_8: KG4CXV2S

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/KG4CXV2S5DJ6NJPIEXYOBBCU33 \
  | 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: 51b82bd752e8d3e6a5e825f0e08454dee89c6d78b54487bb38d8f51d6c89975f

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "fd1d87929fd7b4d269b6c590f982bbd7d4a2c5b5936f370761fecef9ae82ebc9",
    "cross_cats_sorted": [
      "cs.CL"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-02-14T23:45:29Z",
    "title_canon_sha256": "3beb479cc1de2ca9b85214dd4fc028ddb53b3f182fc7c9c928e7dcf24e616ab9"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.04937",
    "kind": "arxiv",
    "version": 1
  }
}