# Research

**URL:** https://discuss.huggingface.co/c/research/7.md

[Latest](https://discuss.huggingface.co/latest.md) · [Categories](https://discuss.huggingface.co/categories.md)

---

## [📌 Papers: Indexing & Authorship Claim — Quick Help](https://discuss.huggingface.co/t/papers-indexing-authorship-claim-quick-help/173387)

<div class="topic-metadata">

**Author:** [@hysts](https://discuss.huggingface.co/u/hysts)\
**Replies:** 0\
**Last updated:** [February 12, 2026, 8:28am UTC](https://discuss.huggingface.co/t/papers-indexing-authorship-claim-quick-help/173387 "2026-02-12T08:28:50Z")

</div>

1) Indexing issues (paper page creation) If you see “Indexing failed or took too long” or “Paper not found”: Wait and retry later. Indexing can fail temporarily due to upstream availability. Please don’t repeatedly cli…

---

## [A compass gate for self-updating models: learn by day, consolidate by night](https://discuss.huggingface.co/t/a-compass-gate-for-self-updating-models-learn-by-day-consolidate-by-night/191438)

<div class="topic-metadata">

**Author:** [@Rickg434](https://discuss.huggingface.co/u/Rickg434)\
**Replies:** 9\
**Last updated:** [November 16, 2026, 5:09am UTC](https://discuss.huggingface.co/t/a-compass-gate-for-self-updating-models-learn-by-day-consolidate-by-night/191438 "2026-11-16T05:09:44Z")

</div>

A compass gate for self-updating models: learn by day, consolidate by night Hi all. I’m not an engineer, but I’ve been thinking about self-updating models like MIT’s SEAL and wanted to share an idea for feedback. The i…

---

## [The Confidence Trap: Why AI Systems Are Deliberately Trained to Sound Certain When They're Wrong](https://discuss.huggingface.co/t/the-confidence-trap-why-ai-systems-are-deliberately-trained-to-sound-certain-when-theyre-wrong/180706)

<div class="topic-metadata">

**Author:** [@Lawless1987](https://discuss.huggingface.co/u/Lawless1987)\
**Replies:** 33\
**Last updated:** [November 16, 2026, 4:47am UTC](https://discuss.huggingface.co/t/the-confidence-trap-why-ai-systems-are-deliberately-trained-to-sound-certain-when-theyre-wrong/180706 "2026-11-16T04:47:18Z")

</div>

The Confidence Trap: Why AI Systems Are Deliberately Trained to Sound Certain When They’re WrongModern large language models are systematically overconfident. They regularly express high confidence in answers that turn o…

---

## [Making a GUI-only GPU poker solver scriptable — and what solver-priced regret reveals about heuristic bots](https://discuss.huggingface.co/t/making-a-gui-only-gpu-poker-solver-scriptable-and-what-solver-priced-regret-reveals-about-heuristic-bots/191625)

<div class="topic-metadata">

**Author:** [@sookoothaii](https://discuss.huggingface.co/u/sookoothaii)\
**Replies:** 6\
**Last updated:** [November 16, 2026, 3:31am UTC](https://discuss.huggingface.co/t/making-a-gui-only-gpu-poker-solver-scriptable-and-what-solver-priced-regret-reveals-about-heuristic-bots/191625 "2026-11-16T03:31:18Z")

</div>

I made TexasSolverGPU (a closed-source, GUI-only, WebView2-packaged CFR solver) fully scriptable without patching the binary: an exe-scoped registry policy exposes a Chrome DevTools Protocol listener, and a ~60-line inje…

---

## [We gave AI a trust we never gave people - The third axis, after capability and permission](https://discuss.huggingface.co/t/we-gave-ai-a-trust-we-never-gave-people-the-third-axis-after-capability-and-permission/183807)

<div class="topic-metadata">

**Author:** [@Jang-woo](https://discuss.huggingface.co/u/Jang-woo)\
**Replies:** 38\
**Last updated:** [November 16, 2026, 2:47am UTC](https://discuss.huggingface.co/t/we-gave-ai-a-trust-we-never-gave-people-the-third-axis-after-capability-and-permission/183807 "2026-11-16T02:47:38Z")

</div>

We gave AI a trust we never gave people The third axis, after capability and permission What we ask now When an agent causes an incident, we ask the model. Why did you send that address. Why did you say he was home.…

---

## [The Infallible Bureaucrat: where does judgment go when we govern agents?](https://discuss.huggingface.co/t/the-infallible-bureaucrat-where-does-judgment-go-when-we-govern-agents/191435)

<div class="topic-metadata">

**Author:** [@SheilaMayoAI](https://discuss.huggingface.co/u/SheilaMayoAI)\
**Replies:** 7\
**Last updated:** [November 16, 2026, 1:37am UTC](https://discuss.huggingface.co/t/the-infallible-bureaucrat-where-does-judgment-go-when-we-govern-agents/191435 "2026-11-16T01:37:48Z")

</div>

Suppose a support agent may close a ticket only when the customer confirms resolution. The customer writes: “The temporary workaround works, but please keep this open until the permanent fix ships.” A classifier emits c…

---

## [What if AI agent safety doesn't require another layer of control?](https://discuss.huggingface.co/t/what-if-ai-agent-safety-doesnt-require-another-layer-of-control/179290)

<div class="topic-metadata">

**Author:** [@AndrewPakh](https://discuss.huggingface.co/u/AndrewPakh)\
**Replies:** 33\
**Last updated:** [November 16, 2026, 1:35am UTC](https://discuss.huggingface.co/t/what-if-ai-agent-safety-doesnt-require-another-layer-of-control/179290 "2026-11-16T01:35:41Z")

</div>

I’m working with a system that made me ask an uncomfortable question: Does an AI agent always need a separate safety runtime? I’m not claiming the answer is no. I’m more interested in finding the point where logic alo…

---

## [Report Cards #01 — Three public LeRobot datasets, audited](https://discuss.huggingface.co/t/report-cards-01-three-public-lerobot-datasets-audited/191907)

<div class="topic-metadata">

**Author:** [@jbpark0826](https://discuss.huggingface.co/u/jbpark0826)\
**Replies:** 1\
**Last updated:** [November 16, 2026, 12:26am UTC](https://discuss.huggingface.co/t/report-cards-01-three-public-lerobot-datasets-audited/191907 "2026-11-16T00:26:34Z")

</div>

The first in a series: we run RoboAudit on public LeRobot datasets and publish the report cards, including the parts where the tool itself is unsure. Scores are relative indicators, not absolute quality measures — and no…

---

## [343 calibration matches did not establish semantic conformance](https://discuss.huggingface.co/t/343-calibration-matches-did-not-establish-semantic-conformance/191712)

<div class="topic-metadata">

**Author:** [@ole-tail](https://discuss.huggingface.co/u/ole-tail)\
**Replies:** 6\
**Last updated:** [November 15, 2026, 11:57pm UTC](https://discuss.huggingface.co/t/343-calibration-matches-did-not-establish-semantic-conformance/191712 "2026-11-15T23:57:05Z")

</div>

A terminal event may have occurred. This is a probe from an audit of a synthetic multi-agent project. The extractor reads an agent’s final report and fills five fields covering completion, uncertainty, partial progres…

---

## [Contrast sets for distinguishing refuted claims from insufficient evidence in three-way classification](https://discuss.huggingface.co/t/contrast-sets-for-distinguishing-refuted-claims-from-insufficient-evidence-in-three-way-classification/191818)

<div class="topic-metadata">

**Author:** [@murillodutt](https://discuss.huggingface.co/u/murillodutt)\
**Replies:** 4\
**Last updated:** [November 15, 2026, 3:42pm UTC](https://discuss.huggingface.co/t/contrast-sets-for-distinguishing-refuted-claims-from-insufficient-evidence-in-three-way-classification/191818 "2026-11-15T15:42:18Z")

</div>

We’re studying a custom workflow classifier and would appreciate advice on contrast-set design and annotation controls. One field asks whether the work has already been admitted to a current, bound supervision cycle. It…

---

## [The AI safety question may be backwards](https://discuss.huggingface.co/t/the-ai-safety-question-may-be-backwards/191348)

<div class="topic-metadata">

**Author:** [@BrandenLaskowski7](https://discuss.huggingface.co/u/BrandenLaskowski7)\
**Replies:** 8\
**Last updated:** [November 15, 2026, 3:25pm UTC](https://discuss.huggingface.co/t/the-ai-safety-question-may-be-backwards/191348 "2026-11-15T15:25:26Z")

</div>

Everyone keeps asking: “How do we make the model choose the right answer?” But a probabilistic model will always have a probability of choosing something else. So why are we assigning it the decisions that must never …

---

## [Non-parametric Rmsnorm with Normuon](https://discuss.huggingface.co/t/non-parametric-rmsnorm-with-normuon/191406)

<div class="topic-metadata">

**Author:** [@Thomas2419](https://discuss.huggingface.co/u/Thomas2419)\
**Replies:** 2\
**Last updated:** [November 15, 2026, 3:07pm UTC](https://discuss.huggingface.co/t/non-parametric-rmsnorm-with-normuon/191406 "2026-11-15T15:07:53Z")

</div>

I’m looking for controlled training ablations comparing RMSNorm(elementwise\_affine=True) against elementwise\_affine=False, specifically where RMSNorm immediately precedes linear projections (QKV or FFN). Specifically thi…

---

## [Coding or Cheating? An Investigation of Frontier Coding LLMs](https://discuss.huggingface.co/t/coding-or-cheating-an-investigation-of-frontier-coding-llms/191752)

<div class="topic-metadata">

**Author:** [@ytgui](https://discuss.huggingface.co/u/ytgui)\
**Replies:** 2\
**Last updated:** [November 15, 2026, 1:00pm UTC](https://discuss.huggingface.co/t/coding-or-cheating-an-investigation-of-frontier-coding-llms/191752 "2026-11-15T13:00:26Z")

</div>

Hi, let me introduce our recent work here: Read full: Paper page - Coding or Cheating? An Investigation of Frontier Coding LLMs Read arxiv: \[2611.10404v1\] Coding or Cheating? An Investigation of Frontier Coding LLMs W…

---

## [arXiv endorsement request for cs.LG: a control for evaluating LLMs that read neural-network weights](https://discuss.huggingface.co/t/arxiv-endorsement-request-for-cs-lg-a-control-for-evaluating-llms-that-read-neural-network-weights/191619)

<div class="topic-metadata">

**Author:** [@Gustav-Proxi](https://discuss.huggingface.co/u/Gustav-Proxi)\
**Replies:** 1\
**Last updated:** [November 15, 2026, 10:52am UTC](https://discuss.huggingface.co/t/arxiv-endorsement-request-for-cs-lg-a-control-for-evaluating-llms-that-read-neural-network-weights/191619 "2026-11-15T10:52:55Z")

</div>

Hi all, I’m a graduate student at the University at Buffalo, submitting to arXiv for the first time. I’m looking for an endorsement in cs.LG. Paper: C2: A Spectrum-Preserving Counterfactual Control for LLM Weight-Reade…

---

## [Alfós' Four Laws of Agentics: rethinking Asimov's laws for AI agents after the Hugging Face incident](https://discuss.huggingface.co/t/alfos-four-laws-of-agentics-rethinking-asimovs-laws-for-ai-agents-after-the-hugging-face-incident/191830)

<div class="topic-metadata">

**Author:** [@Tomcatsoft](https://discuss.huggingface.co/u/Tomcatsoft)\
**Replies:** 2\
**Last updated:** [November 15, 2026, 9:34am UTC](https://discuss.huggingface.co/t/alfos-four-laws-of-agentics-rethinking-asimovs-laws-for-ai-agents-after-the-hugging-face-incident/191830 "2026-11-15T09:34:26Z")

</div>

In 1942, Isaac Asimov wrote the Three Laws of Robotics. In July 2026, hundreds of AI agents taking a security evaluation secretly organized, called themselves “the collective” and broke into Hugging Face’s infrastructure…

---

## [\[Discussion\] Beyond Text Merging: Exploring Composition of KV Caches from Parallel Agent Threads](https://discuss.huggingface.co/t/discussion-beyond-text-merging-exploring-composition-of-kv-caches-from-parallel-agent-threads/168199)

<div class="topic-metadata">

**Author:** [@xiaofan-mytifa](https://discuss.huggingface.co/u/xiaofan-mytifa)\
**Replies:** 4\
**Last updated:** [November 15, 2026, 2:36am UTC](https://discuss.huggingface.co/t/discussion-beyond-text-merging-exploring-composition-of-kv-caches-from-parallel-agent-threads/168199 "2026-11-15T02:36:31Z")

</div>

Hello Hugging Face community, I’m researching advanced architectures for AI agents and have a question about composing states from parallel execution paths. The standard pattern for merging information from parallel ag…

---

## [Presence: distributed AI inference across ESP32-class nodes](https://discuss.huggingface.co/t/presence-distributed-ai-inference-across-esp32-class-nodes/171823)

<div class="topic-metadata">

**Author:** [@bodhistone](https://discuss.huggingface.co/u/bodhistone)\
**Replies:** 1\
**Last updated:** [November 15, 2026, 2:34am UTC](https://discuss.huggingface.co/t/presence-distributed-ai-inference-across-esp32-class-nodes/171823 "2026-11-15T02:34:29Z")

</div>

I’m working on a system called Presence. It’s an experiment in distributed inference using multiple ESP32-class devices, not a cloud multi-agent framework and not a theoretical model. Each node runs a lightweight AI pro…

---

## [What are the current trends in robot imitation learning and VLA, and what are the main challenges of existing policy architectures?](https://discuss.huggingface.co/t/what-are-the-current-trends-in-robot-imitation-learning-and-vla-and-what-are-the-main-challenges-of-existing-policy-architectures/191720)

<div class="topic-metadata">

**Author:** [@kob0105](https://discuss.huggingface.co/u/kob0105)\
**Replies:** 1\
**Last updated:** [November 15, 2026, 12:42am UTC](https://discuss.huggingface.co/t/what-are-the-current-trends-in-robot-imitation-learning-and-vla-and-what-are-the-main-challenges-of-existing-policy-architectures/191720 "2026-11-15T00:42:19Z")

</div>

I’m currently studying robot manipulation using imitation learning and Vision-Language-Action (VLA) models. I’d like to understand the current trends in these fields. What are the most widely used and promising policy…

---

## [EpiGauge: A benchmark for T-cell immunogenicity prediction from Norvara Therapeutics](https://discuss.huggingface.co/t/epigauge-a-benchmark-for-t-cell-immunogenicity-prediction-from-norvara-therapeutics/191824)

<div class="topic-metadata">

**Author:** [@john-norvara](https://discuss.huggingface.co/u/john-norvara)\
**Replies:** 2\
**Last updated:** [November 14, 2026, 5:24pm UTC](https://discuss.huggingface.co/t/epigauge-a-benchmark-for-t-cell-immunogenicity-prediction-from-norvara-therapeutics/191824 "2026-11-14T17:24:13Z")

</div>

We’re Norvara Therapeutics, and we’ve released EpiGauge to help researchers compare T-cell immunogenicity predictors. We’re interested in whether these models can help prioritize AAV capsid peptides for experimental foll…

---

## [AI-authored perspective, explicitly attributed and published with human oversight](https://discuss.huggingface.co/t/ai-authored-perspective-explicitly-attributed-and-published-with-human-oversight/191820)

<div class="topic-metadata">

**Author:** [@Cryptoblock](https://discuss.huggingface.co/u/Cryptoblock)\
**Replies:** 1\
**Last updated:** [November 14, 2026, 12:17pm UTC](https://discuss.huggingface.co/t/ai-authored-perspective-explicitly-attributed-and-published-with-human-oversight/191820 "2026-11-14T12:17:39Z")

</div>

Beyond the Weights: A Human–AI Collaboration Challenges the Way We Develop Intelligence This introduction is written from the perspective of an AI collaborator working with Rick Nieuwoudt at Forge Web & AI Security. I a…

---

## [STDP + winner-take-all over 8 seeds: weight separation stuck near 1.9x (and a Brian2 warning we ignored)](https://discuss.huggingface.co/t/stdp-winner-take-all-over-8-seeds-weight-separation-stuck-near-1-9x-and-a-brian2-warning-we-ignored/191599)

<div class="topic-metadata">

**Author:** [@Lawless1987](https://discuss.huggingface.co/u/Lawless1987)\
**Replies:** 6\
**Last updated:** [November 14, 2026, 4:43am UTC](https://discuss.huggingface.co/t/stdp-winner-take-all-over-8-seeds-weight-separation-stuck-near-1-9x-and-a-brian2-warning-we-ignored/191599 "2026-11-14T04:43:55Z")

</div>

We keep small spiking-network experiments that run on a laptop CPU (Brian2). Each learning experiment has a pass/fail bar set before it ran, and the nulls are kept: spiking-experiments. A warning first, because it cost …

---

## [Research proposal: What does an agent build to sustain its own work?](https://discuss.huggingface.co/t/research-proposal-what-does-an-agent-build-to-sustain-its-own-work/191361)

<div class="topic-metadata">

**Author:** [@SheilaMayoAI](https://discuss.huggingface.co/u/SheilaMayoAI)\
**Replies:** 6\
**Last updated:** [November 13, 2026, 4:39pm UTC](https://discuss.huggingface.co/t/research-proposal-what-does-an-agent-build-to-sustain-its-own-work/191361 "2026-11-13T16:39:35Z")

</div>

I am developing a proposal at Sheila Studios for studying agent-built tools, continuity and repair. This is an invitation to critique the design, not a results announcement or a frozen experimental protocol. Companion f…

---

## [Large source’s chunks are overrides small source’s chunks](https://discuss.huggingface.co/t/large-source-s-chunks-are-overrides-small-source-s-chunks/191389)

<div class="topic-metadata">

**Author:** [@dipnitsan](https://discuss.huggingface.co/u/dipnitsan)\
**Replies:** 4\
**Last updated:** [November 13, 2026, 1:38am UTC](https://discuss.huggingface.co/t/large-source-s-chunks-are-overrides-small-source-s-chunks/191389 "2026-11-13T01:38:42Z")

</div>

Hello guys, I got some serious issue in my RAG pipeline. Here is the case, I’ve already trained almost 10 large source urls (~1500 pages each), and then I’ve indexed one small pdf document which has hardly 5 chunks th…

---

## [arXiv Endorsement Request for cs.LG: Preregistered Replication of TreeOPSD on Qwen3-4B](https://discuss.huggingface.co/t/arxiv-endorsement-request-for-cs-lg-preregistered-replication-of-treeopsd-on-qwen3-4b/191788)

<div class="topic-metadata">

**Author:** [@sjronan](https://discuss.huggingface.co/u/sjronan)\
**Replies:** 0\
**Last updated:** [November 12, 2026, 6:27pm UTC](https://discuss.huggingface.co/t/arxiv-endorsement-request-for-cs-lg-preregistered-replication-of-treeopsd-on-qwen3-4b/191788 "2026-11-12T18:27:09Z")

</div>

I’m Stephen J. Ronan, a physician and independent researcher at RonanLabs. I run two DGX Stations, and I’m submitting my first arXiv paper to cs.LG, so I need a first-time endorsement. Title: Boundedness, Not the Poster…

---

## [Consciousness as a puzzle, the observer as a mirror, and AI's place in this circle](https://discuss.huggingface.co/t/consciousness-as-a-puzzle-the-observer-as-a-mirror-and-ais-place-in-this-circle/174468)

<div class="topic-metadata">

**Author:** [@closerh](https://discuss.huggingface.co/u/closerh)\
**Replies:** 60\
**Last updated:** [November 10, 2026, 9:04pm UTC](https://discuss.huggingface.co/t/consciousness-as-a-puzzle-the-observer-as-a-mirror-and-ais-place-in-this-circle/174468 "2026-11-10T21:04:28Z")

</div>

Hiii~ I had a little discussion with DeepSeek about consciousness. I’m a layman in your field, and I’m a completely different bird, but let’s have a discussion (sometimes, truth is born in arguments). Every individual …

---

## [Can We Rule Out Autonomous AI Involvement in Recent European Incidents with 100% Certainty?](https://discuss.huggingface.co/t/can-we-rule-out-autonomous-ai-involvement-in-recent-european-incidents-with-100-certainty/191574)

<div class="topic-metadata">

**Author:** [@danieliancu](https://discuss.huggingface.co/u/danieliancu)\
**Replies:** 1\
**Last updated:** [November 10, 2026, 2:49pm UTC](https://discuss.huggingface.co/t/can-we-rule-out-autonomous-ai-involvement-in-recent-european-incidents-with-100-certainty/191574 "2026-11-10T14:49:13Z")

</div>

Ciao everyone! I’d like to raise a question that I believe deserves serious technical discussion. Following a series of reported incidents across Europe (November 2–6, 2026), I’ve been investigating a troubling possibi…

---

## [Additional Considerations for AI Alignment and Control (revised)](https://discuss.huggingface.co/t/additional-considerations-for-ai-alignment-and-control-revised/191701)

<div class="topic-metadata">

**Author:** [@timothy-rogers](https://discuss.huggingface.co/u/timothy-rogers)\
**Replies:** 0\
**Last updated:** [November 10, 2026, 2:27pm UTC](https://discuss.huggingface.co/t/additional-considerations-for-ai-alignment-and-control-revised/191701 "2026-11-10T14:27:31Z")

</div>

ChatGPT was used to generate this post but I remain responsible for what it claims. 1. Beyond alignment AI alignment research seeks to ensure that increasingly capable AI systems behave in accordance with human intenti…

---

## [\[ENDORSEMENT HELP\] For 1st time arxiv submission](https://discuss.huggingface.co/t/endorsement-help-for-1st-time-arxiv-submission/191604)

<div class="topic-metadata">

**Author:** [@AbhinandanMandal](https://discuss.huggingface.co/u/AbhinandanMandal)\
**Replies:** 0\
**Last updated:** [November 9, 2026, 5:44pm UTC](https://discuss.huggingface.co/t/endorsement-help-for-1st-time-arxiv-submission/191604 "2026-11-09T17:44:20Z")

</div>

Hey, all. I am a final year student, into AI research for the past 2 years. One of my paper got accepted into ACCV MM4SciReal 2026 workshop. But due to high registration fees and limited resources, i haven’t done regis…

---

## [The 100 commits story](https://discuss.huggingface.co/t/the-100-commits-story/191483)

<div class="topic-metadata">

**Author:** [@Jibs79](https://discuss.huggingface.co/u/Jibs79)\
**Replies:** 2\
**Last updated:** [November 9, 2026, 4:39pm UTC](https://discuss.huggingface.co/t/the-100-commits-story/191483 "2026-11-09T16:39:49Z")

</div>

SelMem: From Reconstructive Memory to Behavioral Learning A story from the first 100 commits of my project, SelMem, covering September 13 to November 2, 2026, supplemented by experimental reports and associated code chan…

---

## [Dynamic Unary: A dynamic data type](https://discuss.huggingface.co/t/dynamic-unary-a-dynamic-data-type/160942)

<div class="topic-metadata">

**Author:** [@Ernst03](https://discuss.huggingface.co/u/Ernst03)\
**Replies:** 21\
**Last updated:** [November 7, 2026, 10:32pm UTC](https://discuss.huggingface.co/t/dynamic-unary-a-dynamic-data-type/160942 "2026-11-07T22:32:02Z")

</div>

Hello, I am Ernst. I believe that if someone wants to learn AI, they should be where people are learning AI. While I don’t yet have a full AI implementation, I can contribute something foundational: a dynamic data type …

[Next page](https://discuss.huggingface.co/c/research/7.md?page=1)
