Skip to content
View lloitesa013's full-sized avatar

Block or report lloitesa013

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
lloitesa013/README.md

Dana Kim · 김단아

Independent Researcher · @lloitesa013

Contact: lloitesa@gmail.com · OpenReview: ~Dana_Kim1

Research Kit Verification CI

I build the layer that decides when an AI system's output can be trusted — and what happens when it can't: a runtime watcher that abstains and escalates, failure taxonomies that localize why models fail, and a verifier that gates what a self-improving agent may adopt.

AI를 더 똑똑하게 만드는 대신, AI의 출력을 언제 믿어도 되는지 — 믿으면 안 될 때 무엇을 해야 하는지 — 를 결정하는 레이어를 만듭니다: abstain/escalate하는 런타임 감시자(watcher), 실패 원인을 국소화하는 택소노미, 자기개선 에이전트의 수정을 게이트하는 검증기.

I do not build more impressive AI outputs; I build the evidence-bound judgment above them.

Three doors

Looking for… Go to The one number
Agent reliability / hallucination control (production) agent-reliability-bench — Layer 1 watcher: trust score, ABSTAIN, ESCALATE, audit log. Offline tour: python demo.py zero-shot reading judge 0.946±0.014 vs 0.789 tuned groundedness rule, 0 unsafe passes (sealed 300-trace bench); abstention lifts 0.84 → 0.91
Driving evaluation & failure analysis (research) bench2drive-failure-taxonomy + xmod-interpretable-driving LEAD/TransFuser-V6 DS 93.36 reproduced over 220 routes + per-route failure taxonomy, expert-calibrated; X-MoD = an honest, claims-bounded closed-loop study
Verified self-improvement (frontier) agent-reliability-bench — Layer 2 VSI-0: hidden-test + regression gate over an agent's own {prompt, rule, memory, tool} changes naive self-modification ships a broken tool 59% of the time, the executing verifier 0% (24 decisions; replicated on MBPP + HumanEval; embodied ACCEPT on CARLA, p=6e-8)

Built in public — 10 weeks, one RTX 5090

2026 Shipped
May 27–31 Financial Agent Evidence OS — claim contracts + release gates (early iteration of the thesis)
May 28 – Jun 1 Cognitive OS API — rule-based decision gate ALLOW/DEGRADE/DENY/HANDOFF (early iteration)
Jun 12–19 Bench2Drive failure taxonomy — 220-route reproduction (DS 93.36) + first per-route failure map with expert calibration
Jun 14 Mantle audit devkit — the same evidence-packet pattern applied to smart contracts (hackathon transfer test)
Jun 22 Merged upstream: autonomousvision/carla_garage PR #117 — Bench2Drive integration fix for numpy ≥ 1.24
Jun 25 X-MoD — interpretable driving policy; bounded negative result published as such, with paper
Jul 1–9 agent-reliability-bench — the watcher (sealed benchmark, R1–R7) + VSI-0 verified self-improvement (R8–R31)

Final Standard

Works. Reproducible. Measured. Understandable. Bounded.

The public portfolio should not rely on "I am impressive" as a claim. It should show running code, reproducible evidence, measurable gates, clear explanations, and visible limits.

Read: Verification Principles Role map: Role Positioning

Proof at a Glance

Work It works Reproducible Measured Boundary
Agent-Reliability Watcher × VSI-0 python demo.py — full offline watcher tour, no key; vsi_gate.py = the verifier as a pure-stdlib library Sealed spec (BENCH_SPEC/METRICS/CLAIMS, win conditions fixed before results) + 300-trace data + every result names its script Judge 0.946±0.014 vs rule 0.789, 0 unsafe passes; abstention 0.84→0.91; naive ships broken self-mods 59% vs verifier 0% (MBPP/HumanEval replicated) Synthetic bench, 7B models, single GPU; not production, not SOTA; honest negatives are first-class results
Bench2Drive Failure Taxonomy Reproducible SOTA baseline plus per-route failure map Two harnesses, per-route evidence (~400 raw route JSONs), calibration tooling LEAD/TransFuser-V6 DS 93.36 vs PDM-Lite 95.71; failures split fixable / structural / infra via privileged expert Single-GPU/seed; not a new SOTA; per-route labels refined to failure rates by follow-up (R19–R21)
X-MoD on Bench2Drive Closed-loop B2D agent, paper, eval harness Code plus 3-seed eval protocol, claims/non-claims Unseen route 24330: 100% completion, 0 collisions across 3 seeds; gate/safety corr 0.80 Does not generalize uniformly; completion uses a classical control stack
Cognitive OS API FastAPI evidence viewer and /evidence/report Seed benchmark, baselines, conformance, adversarial runner Seed-suite gate accuracy 100% (authored against the rule set); external-style challenge pack 37.5% with failure analysis — the honest number Rule-based early prototype; not AGI, not global LLM safety
Financial Agent Evidence OS Streamlit evidence viewer and release gates Full unit suite plus release/audit/verifier commands 118 tests OK, 2 skipped; unfavorable market results preserved in the record Not financial advice, not live trading, not future-return prediction
Angelos Reproducibility Kit Public packet validator Frozen anchors and manifest checks ped_surprise/v10, merge_cutin/v11, 40 seeds each CARLA protocol packet, not deployment readiness

X-MoD drive alignment status

Angelos pedestrian surprise schematic

Flagship — Agent-Reliability Watcher × VSI-0

One system, two layers, one question: when should an AI agent's output — or its own self-modification — be trusted?

  • Layer 1, the watcher (runtime): groundedness × relevance trust scoring, ABSTAIN when unsure, ESCALATE to a human with a reason, audit log for every decision. Evidence: sealed 300-trace benchmark, judge 0.946±0.014 vs tuned rule 0.789, 0 unsafe passes.
  • Layer 2, VSI-0 (research): every self-modification is UNTRUSTED until it clears a hidden test + regression check — because reward-hacked "improvements" are the norm, not the exception (59% naive deployment rate of broken tools; verifier 0%).
  • The two layers are one thesis at two levels of autonomy: watch the outputs, then gate the self-modifications.

Repo: agent-reliability-bench

Earlier iterations (kept public, honestly labeled)

  • Cognitive OS API — 2026-05 rule-based decision gate (ALLOW/DEGRADE/DENY/HANDOFF): the HANDOFF-to-human primitive that became the watcher's ESCALATE.
  • Financial Agent Evidence OS — 2026-05 claim contracts + sealed evidence packets for a financial agent, negative results preserved.
  • Mantle audit devkit — the same proof-carrying-evidence pattern transferred to smart-contract audits (hackathon).

Research Lineage

My systems grew out of a single question:

How can an AI system decide, restrain, explain, and verify its actions under explicit human values?

  1. Flow-Based AI Ethics Model: ethical cognition as a decision flow.
  2. X-MoD: explainable routing over explicit driving values.
  3. Angelos OS: safety-gated judgment above autonomous driving policies.
  4. Cognitive OS: user-owned policy and decision gates above LLMs.
  5. Evidence OS: benchmark and release gates for claims made by agents.
  6. Agent-Reliability Watcher × VSI-0: the judgment layer at two levels of autonomy — watch the agent's outputs, then gate the agent's self-modifications.

Read more: AI Judgment Lineage

Autonomous Driving Research Line

X-MoD and Angelos OS are the research roots of this portfolio. They test the same judgment-layer idea in a harder setting: autonomous driving, where action, restraint, explanation, and failure analysis must be tied to scenario evidence.

  • X-MoD: explainable routing over safety, legality, comfort, and efficiency.
  • Angelos OS / SynOptic Core: a model-agnostic judgment layer above driving policies, with safety gates and protocol-frozen evaluation.
  • Public repos: X-MoD on Bench2Drive (closed-loop interpretable driving, paper, and eval harness) and Bench2Drive Failure Taxonomy (reproducible SOTA baseline plus per-route failure map); plus the offline X-MoD Visual Paper Kit and Angelos Reproducibility Kit.
  • Upstream: carla_garage PR #117 (Bench2Drive numpy ≥ 1.24 fix, merged 2026-06-22).
  • Boundary: these kits are sanitized research packets, not broad safety or deployment claims. The follow-up reliability work showed single-run per-route failure labels flip on re-run — so claims are stated at failure-rate level (agent-reliability-bench R19–R21).

Portfolio Map

Proof Surface

Python - PyTorch/transformers - CARLA/Bench2Drive - FastAPI - Streamlit - CI gates - sealed benchmark specs - evidence manifests - release gates - decision envelopes - trace privacy

Claim Boundary

This portfolio does not claim external adoption, global SOTA status, AGI, live trading readiness, investment performance, complete AI safety, or any unbounded guarantee. Claims are scoped to the included reference architectures, benchmark suites, and published evidence artifacts.

Pinned Loading

  1. cognitive-os-api cognitive-os-api Public

    2026-05 iteration of the verification thesis: a rule-based LLM decision gate (ALLOW/DEGRADE/DENY/HANDOFF) with evidence packets and honestly reported challenge failures — early prototype of the age…

    Python

  2. stock-agent-harness-v0-2-0-defense stock-agent-harness-v0-2-0-defense Public

    2026-05 iteration of the verification thesis in finance: claim contracts + release gates + sealed evidence packets for a financial agent — including preserved negative results (weak real-market per…

    Python

  3. bench2drive-failure-taxonomy bench2drive-failure-taxonomy Public

    Reproducible LEAD/TransFuser-V6 Bench2Drive-220 baseline (DS 93.36) + first public per-route failure taxonomy of a SOTA open driving model

    Python

  4. xmod-interpretable-driving xmod-interpretable-driving Public

    Interpretable Mixture-of-Drives on Bench2Drive: a transparent driving policy (safety gate corr 0.80) that completes an unseen route 100% (3-seed robust) but does not generalize uniformly. Honest, c…

    Python