Pith Number

pith:HA273CXN

pith:2025:HA273CXNLRK6Z3ZEUNKIWPUO2O

not attested not anchored not stored refs pending

Tapered Off-Policy REINFORCE: Stable and efficient reinforcement learning for LLMs

Alex Fr\'echette, Arnaud Bergeron, Carolyne Pelletier, Eric Thibodeau-Laufer, Jonathan Lebensold, Joshua Greaves, Marc G. Bellemare, Nicolas Le Roux, Sam Work, S\'andor Toth

arxiv:2503.14286 v2 · 2025-03-18 · cs.LG

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{HA273CXNLRK6Z3ZEUNKIWPUO2O}

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.

Cited by

14 papers in Pith

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma

What are Key Factors for Updates in RL for LLM Reasoning?

Modularized Reinforcement Learning on LLMs: From MDP Creation to Exploration and Learning

Rollout-Level Advantage-Prioritized Experience Replay for GRPO

Trust Region On-Policy Distillation

Receipt and verification

First computed	2026-07-05T10:34:43.804692Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

3835fd8aed5c55ecef24a3548b3e8ed38a7c47f9e50f1ce6f266ec683b7c631b

Aliases

arxiv: 2503.14286 · arxiv_version: 2503.14286v2 · doi: 10.48550/arxiv.2503.14286 · pith_short_12: HA273CXNLRK6 · pith_short_16: HA273CXNLRK6Z3ZE · pith_short_8: HA273CXN

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/HA273CXNLRK6Z3ZEUNKIWPUO2O \
  | 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: 3835fd8aed5c55ecef24a3548b3e8ed38a7c47f9e50f1ce6f266ec683b7c631b

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "170c1475e304016e31622d2bc8f51339fe403395782174af126da8e0c1c24042",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2025-03-18T14:23:37Z",
    "title_canon_sha256": "289d6d612117f3e318153231d480fc23b6df73de45ce703d7e21c3097f5704f6"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2503.14286",
    "kind": "arxiv",
    "version": 2
  }
}