Pith Number

pith:XFSFPFTM

pith:2026:XFSFPFTMQIZ4KNIJGBZQUFDFA2

not attested not anchored not stored refs resolved

Constrained Code Generation with Discrete Diffusion

Ferdinando Fioretto, Lize Shao, Michael Cardei, Wenxi Wang, Zichen Xie

Constrained Diffusion for Code augments discrete diffusion samplers with optimization-driven operators that steer denoising toward programs satisfying functional, security, and syntax constraints.

arxiv:2605.16829 v1 · 2026-05-16 · cs.CL · cs.PL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{XFSFPFTMQIZ4KNIJGBZQUFDFA2}

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

CDC augments the base discrete diffusion sampler with constraint-aware denoising operators that combine mathematical optimization with program analysis to identify constraint-relevant regions of the intermediate program state and locally adjust the denoising trajectory, steering generation toward feasible programs while remaining close to the base model.

C2weakest assumption

The assumption that program analysis can reliably identify constraint-relevant regions in noisy or partially denoised intermediate program states and that local adjustments via optimization will steer the trajectory toward satisfying constraints without harming overall sample quality or requiring model retraining.

C3one line summary

Constrained Diffusion for Code (CDC) integrates constraint satisfaction into the reverse denoising process of discrete diffusion models via constraint-aware operators that use optimization and program analysis to steer generation toward feasible programs.

References

36 extracted · 36 resolved · 11 Pith anchors

[1] Mankowitz, Esme Sutherland Robson, Pushmeet Kohli, Nando de Freitas, Koray Kavukcuoglu, and Oriol Vinyals 2022 · doi:10.1126/science.abq1158

[2] arXiv preprint arXiv:2303.12570 , year= 2023

[3] SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering 2024 · arXiv:2405.15793

[4] Diffucoder: Understanding and improving masked diffusion models for code generation 2025

[5] arXiv preprint arXiv:2509.01142 , year= 2025

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-20T00:03:24.858095Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

b96457966c8233c5350930730a146506aafe593101a5b28afcbc7b7641c415b0

Aliases

arxiv: 2605.16829 · arxiv_version: 2605.16829v1 · doi: 10.48550/arxiv.2605.16829 · pith_short_12: XFSFPFTMQIZ4 · pith_short_16: XFSFPFTMQIZ4KNIJ · pith_short_8: XFSFPFTM

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/XFSFPFTMQIZ4KNIJGBZQUFDFA2 \
  | 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: b96457966c8233c5350930730a146506aafe593101a5b28afcbc7b7641c415b0

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "45ab85a239876e39036ce05ab114c3b3d8b16f30ded283362d12db147abc5221",
    "cross_cats_sorted": [
      "cs.PL"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-05-16T06:15:47Z",
    "title_canon_sha256": "3e5b7fcb567be5220e2a5b0254734beef73a1ff7cb0d835c20d7e2de232a020a"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.16829",
    "kind": "arxiv",
    "version": 1
  }
}