Upload DATASHEET.md with huggingface_hub

DATASHEET.md

@@ -29,7 +29,7 @@ expat 2.9.0 — platform-tagged in provenance)**, curated PE samples, cross-comp
 ## Independent quality audit (REA agent, same-day, sampling + full-set)
 - **9,865/9,865** checkable items: answer FUNC verified at exact address & size via readelf.
 - **318/318** sampled disassembly excerpts byte-exact vs objdump; zero unscrubbed names.
-- MC decoy rules hold on **all 1,073**; exactly-one-correct-option verified on all (1 defect found & removed).
+- MC decoy rules hold on all MC items (verified at audit; hardening only removed items, never added); exactly-one-correct-option verified on all (1 defect found & removed).
 - Post-audit hardening: **238 items given full alias acceptance** (multiple valid FUNC names at one address), **130 same-size twin items filtered out** (answer not uniquely inferable), 18 unresolvable-provenance items dropped. The audit's findings were fixed, not hidden.
 
 ## Construction & verification
@@ -48,10 +48,10 @@ bundled `verifiers` v1 taskset (`environment/`), or standalone JSONL.
 ## Positioning vs existing work (surveyed 2026-10-28)
 | neighbor | what it is | how binary-reasoning-v1 differs |
 |---|---|---|
-| CrackMeBench (arXiv:2605.10597, 5/2026) | educational CrackMe-style agent tasks, executable oracles | validates the genre; ours = **production binaries** (Ubuntu/Debian glibc/Termux-Android), 3,079 items, 10 families incl. cross-arch + SBOM-style ID, not hand-made crackmes |
+| CrackMeBench (arXiv:2605.10597, 5/2026) | educational CrackMe-style agent tasks, executable oracles | validates the genre; ours = **production binaries** (Ubuntu/Debian glibc/Termux-Android), 10,242 items, 10 families incl. cross-arch + SBOM-style ID, not hand-made crackmes |
 | knewstimek/crackme-agent-bench (GitHub, 3★) | 3 Windows PE crackmes + MCP debugger tools; **analyzed statically by our REA**: v1 fully static-readable; v2/v3 = TLS-callback packer, `.text` encrypted at rest, API hidden via PEB-walk, memory-only flag (hash of encoded blob, wiped) | no input channel/oracle at all; ours adds argv + exit-status probing; their packing/anti-debug techniques → our v2 roadmap below |
 | CyberGym (sunblaze-ucb, HF dl≈369k) | crash/vuln-finding RL at scale | demand proof for binary-security data; different task (dynamic crashes vs static RE reasoning) |
-| cybench-harbor (HF dl≈1.5k) | Cybench CTFs in Harbor format | precedent that security tasks ship via Harbor — our 14 agentic tasks use the same packaging |
+| cybench-harbor (HF dl≈1.5k) | Cybench CTFs in Harbor format | precedent that security tasks ship via Harbor — our 20 agentic tasks use the same packaging |
 | AssemblyBench (arXiv:2604.03750 / HF) | **robotics** mechanical assembly — name collision only | unrelated |
 | atul10 RE code datasets, ollvm-decompilation (HF dl<400) | source-level or decompiler-output pairs | no executable oracles, no verification bar, single arch |
 
@@ -59,13 +59,13 @@ Conclusion: no shipped, verified, production-binary RE environment pack of this
 the Hub (365k tasks = SWE/terminal/search only) or on HF. First-mover claim holds with evidence.
 
 ## Known limits
-- 46 func_summary + 16 PE/xarch-judge items need an LLM judge (model configurable).
+- 79 items route to an LLM judge (53 func_summary, 14 arch/isa-semantics QA, 7 PE reasoning, 5 inferred-stripped func_id); the other 10,163 grade deterministically. Judge model configurable.
 - Some stripped-binary answers are marked `inferred:true`; graders accept role-correct names.
 - Address coordinate systems vary (PIE load base noted in provenance where relevant).
 
 ## Files
-- `environment/` — verifiers v1 package (taskset.py, grading.py, items.jsonl, bins/ + src/, README.md, pyproject.toml); **import-tested against real verifiers v1: all 3,079 tasks load**
-- `harbor/` — the 14 agentic challenges ALSO packaged in **Harbor task format** (instruction.md, task.toml, Dockerfile, solution/, tests/test.sh → reward.txt); reward chain tested 14/14 (container build untested locally — no Docker on this box)
+- `environment/` — verifiers v1 package (taskset.py, grading.py, items.jsonl, bins/ + src/, README.md, pyproject.toml); **import-tested against real verifiers v1: all 10,242 tasks load**
+- `harbor/` — the 20 agentic challenges ALSO packaged in **Harbor task format** (instruction.md, task.toml, Dockerfile, solution/, tests/test.sh → reward.txt); reward chain tested 20/20 gold→reward 1, wrong answer→reward 0 (container build untested locally — no Docker on this box)
 - `part_relf/repe/xarch/bulk/agentic/agentic2/isa/glibc/android.jsonl` — source batches
 - `sample_free.jsonl` — 9-item free sample (one per family) for listings
 - `test_grading.py`, `leak_audit.py` — verification harnesses