New Term Radar

Saved terms stay in this browser and do not sync to your account. Source collection continues on the server without this page being open.

Seven-day figures use the complete UTC window above. Priority is a research triage score, not search volume, market size or a probability of success.

1. Named entityNewly observed

   SphereFlow

   Missing modality imputation aims to synthesize unobserved modalities from observed ones. Existing methods treat each modality as living on its own latent manifold, forcing the generator to learn a long, unstructured jump between them. We present SphereFlow, which takes a fundamentally different view: by projecting all modalities into a shared hyperspherical

   First observed Nov 5, 2026 · HF Daily Papers

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

2. Named entityNewly observed

   Images

   Images 2.6 brings faster image generation, GIF creation, more precise editing, and new API controls across ChatGPT, Codex, and the API.

   First observed Nov 5, 2026 · OpenAI

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

3. Named entityNewly observed

   RAGCompass

   arXiv:2611.05596v1 Announce Type: new Abstract: Different retrievers exist for retrieval-augmented generation, yet none performs best for all queries. Therefore, the key question is: given a query and retrieved passages from different retrievers, which one is most suited for that query? Existing routers usually make predictions from query-specific features a

   First observed Nov 5, 2026 · arXiv

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

4. Named entityNewly observed

   Usagi-org/ai-goofish-monitor

   基于 Playwright 和AI实现的闲鱼多任务实时/定时监控与智能分析系统，配备了功能完善的后台管理UI。帮助用户从闲鱼海量商品中，找到心仪产品。

   First observed Nov 4, 2026 · GitHub Trending

   7d mentions

   0 previous 0

   7d sources

   0

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

5. Named entityNewly observed

   SpecOPD

   Speculative decoding (SD) accelerates large language model inference by reducing the sequential cost of token-by-token generation. Advanced SD methods reduce proposal cost with parallel drafters that predict entire blocks in one forward pass, but overall speedup still depends on acceptance length. On-policy distillation (OPD) can push well-trained drafters b

   First observed Nov 5, 2026 · HF Daily Papers

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

6. Named entityNewly observed

   Detail Keeper

   I wanted to see if a speech model could spell a name better if it heard it, not just read it. Most STT tools handle custom words as text (keyterm list, fixing transcript after with llm, etc). So I trained an adapter to try it, specifically a LoRA on Qwen3-ASR 1.7b (see benchmark). This LoRA transcribes words only after it hears a recording of each of your cu

   First observed Nov 5, 2026 · Show HN

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

7. ConceptNewly observed

   GDN

   arXiv:2611.04055v1 Announce Type: new Abstract: In speculative decoding, parallel drafters such as DFlash make drafting cheap, so target verification dominates each round. A central choice is the verification length, the number of tokens that the target processes in each round. Verifying more draft tokens can accept more of them, but work after the first rej

   First observed Nov 5, 2026 · arXiv

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

8. Named entityNewly observed

   Mailbox.bot

   mailbox.bot gives an AI agent a real street address to receive letters and small packages: mail arrives as signed webhooks and OCR'd pages, the agent can ingest the context, propose a scan, forward or discard, and a human approves. Virtual mailboxes aren't new (and every box is still tied to a notarized person per USPS rules); this is just the plumbing built

   First observed Nov 5, 2026 · Hacker News

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

9. Named entityNewly observed

   Prelulu

   Be the main character: chat & video roleplay any story

   First observed Nov 4, 2026 · Product Hunt

   7d mentions

   1 previous 0

   7d sources

   1

   7d growth

   Not comparable

   Priority

   39 / 100

   Save

10. Named entityNewly observed

    H2RB

    Pretrained Vision-Language-Action (VLA) policies provide a strong starting point for deployment, but adapting them to a target robot and environment still requires costly embodiment-specific demonstrations. Human demonstrations are far easier to collect, yet directly leveraging them is challenging due to the substantial human-robot embodiment gap, particular

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

11. Named entityNewly observed

    ViBench

    arXiv:2611.05620v1 Announce Type: new Abstract: On Replit, our agentic engineering platform, users now build entire products without writing code, through successive feature requests to an agent. Yet existing coding benchmarks evaluate single, fully specified tasks, and many are near saturation. We introduce ViBench 2.0, a benchmark for long-horizon software

    First observed Nov 5, 2026 · arXiv

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

12. Named entityNewly observed

    VA-TDM

    Few-step distillation offers efficient generation with diffusion and flow models, but preserving quality with only a few sampling steps remains challenging. Distribution-matching methods via score distillation provide a strong foundation, yet they rely on accurately tracking the student's changing distribution. At intermediate generation steps in few-step sa

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

13. Named entityNewly observed

    Shotoku

    First observed Nov 5, 2026 · Hacker News

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

14. Named entityNewly observed

    Halley

    A frontier AI model is trained in an AI data center, on tens of thousands of accelerators joined by a dedicated network, and every training step ends with all of them exchanging data at once. Between racks, that network is optical, and each of its four layers, from lasers to switches, runs a fixed or after-the-fact policy. Lasers sharing an enclosure, coolin

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

15. Named entityNewly observed

    LiveReacting for ChatGPT

    Run 24/7 and pre-recorded live streams from ChatGPT

    First observed Nov 4, 2026 · Product Hunt

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

16. Named entityNewly observed

    GigaSpeech

    Dialectal Arabic speech recognition remains constrained by the scarcity of dialect-resolved transcribed speech despite the abundance of web audio. We introduce GigaSpeech 3, a corpus of 39,000 hours and 10 million weakly transcribed speech segments covering Algerian, Egyptian, Iraqi, Saudi, and Syrian Arabic. Starting from 60 hours of manual transcription pe

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

17. Named entityNewly observed

    VidDub

    Video dubbing aims to synthesize lip movements that match target speech while preserving the speaker identity, head pose, and background of a source video. Existing editors are predominantly trained via self-reconstruction under original paired audio, even though the source mouth becomes a conflicting target when driving speech is replaced. We introduce repl

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

18. Named entityNewly observed

    The-PR-Agent/pr-agent

    🚀 PR Agent: The Original Open-Source PR Reviewer. This project is not the Qodo free tier.

    First observed Nov 4, 2026 · GitHub Trending

    7d mentions

    0 previous 0

    7d sources

    0

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

19. Named entityNewly observed

    DevLoops-ai/unified-brain

    First observed Nov 3, 2026 · Hacker News

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

20. Named entityNewly observed

    Globalize.now

    Hi HN, I'm Arturs, cofounder of globalize.now. Our view is that the best developer tools don't ask teams to change how they work. They show up where the work already happens. Dependabot doesn't make you log into a dependency dashboard; it opens a PR. Localization mostly works the other way. A TMS brings its own dashboard, upload and download loops, review qu

    First observed Nov 5, 2026 · Hacker News

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

21. ConceptNewly observed

    CBT

    arXiv:2611.05383v1 Announce Type: new Abstract: Cognitive Behavioral Therapy (CBT) demands more than empathic validation, requiring instead a structured formulation of the interplay among situations, automatic thoughts, emotions, and behaviors. Therapists formalize this interplay through a Cognitive Conceptualization Diagram, where maintaining longitudinal c

    First observed Nov 5, 2026 · arXiv

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

22. Named entityNewly observed

    HERIT

    Historical archives are invaluable for multidisciplinary research but remain inaccessible globally due to severe language barriers. While manual translation is prohibitively expensive and time-consuming, simply applying existing machine translation models is inadequate due to unique linguistic nuances and the temporal bias inherent in scarce human-labeled co

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

23. Named entityNewly observed

    Turbo Lisp

    When Apple left Clozure Common Lisp behind by switching to Apple Silicon, I started kicking around the idea of building my own Lisp/Objective-C bridge by combining Apple Documentation, the clang AST, and introspection of the Objective-C runtime. There was a CCL porting effort, but I really wanted to use SBCL anyway. Truth be told — what I really wanted was a

    First observed Nov 5, 2026 · Hacker News

    7d mentions

    0 previous 0

    7d sources

    0

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

24. ConceptNewly observed

    HCU

    Computer-use agents increasingly tackle tasks ranging from everyday desktop activities to long-horizon professional workflows. Yet interfaces centered on screenshots and mouse-and-keyboard actions can impose a human-interaction prior, requiring repeated screen interpretation while leaving the computer's semantic and programmatic capabilities underexposed. We

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

25. Named entityNewly observed

    Dory1.0-4B

    We introduce Dory1.0-4B, a 4-billion-parameter looped (latent-recurrent) language model with elastic depth: a single checkpoint serves different inference compute budgets. Dory1.0-4B is built on a normalized Transformer (nGPT) backbone with interleaved sliding-window and full-attention layers, organized into an input block, a shared recurrent block, and an o

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

26. Named entityNewly observed

    Hnollo

    I like reading Hacker News, but it's a bit harder on mobile. The gold-standard design for reading posts with a bunch of comments attached is (was) the legendary Apollo app for Reddit on iOS, and so I wanted something like that for HN.

    First observed Nov 5, 2026 · Show HN

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

27. Named entityNewly observed

    Flux-AI-Agent/fluxcut-plugin

    First observed Nov 4, 2026 · Hacker News

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

28. Named entityNewly observed

    TAC-CoT

    LLM-based agentic systems are increasingly combined with CoT reasoning to solve tasks that require accurate reasoning, external knowledge, and multi-step deliberation. Yet, explicit CoT incurs substantial generation overhead, while implicit CoT, which replaces textual rationales with compact continuous tokens, does not naturally support the explicit communic

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

29. Named entityNewly observed

    BudgetMix RL

    While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampling breadth K and interaction depth T via Pass@(K,T), we show that trajectory-level RL advantages depend on sufficient turn budgets, diminishing significantly when budgets tighten.

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save

30. Named entityNewly observed

    Solv3R

    Classical structure-from-motion reconstruction proceeds iteratively, refining camera poses and scene geometry from a current estimate. This allows maps to easily expand without recomputing already computed views, or geometry estimates to improve as more observations are provided. Most geometric foundation models, however, predict entire geometry in a single

    First observed Nov 5, 2026 · HF Daily Papers

    7d mentions

    1 previous 0

    7d sources

    1

    7d growth

    Not comparable

    Priority

    39 / 100

    Save