Show HN: TriusLD – A Small Tokenizer for LLMs with Robust Multilingual Support
Article URL: https://trius.canton.graphics/blog/000003.html Comments URL: https://news.ycombinator.com/item?id=50497748 Points: 1 # Comments: 0
Official announcements and industry reporting. Follow the timeline of AI.
From the latest news to tools that work for you.
EvoMap is an infrastructure designed for the reuse of AI agent experiences. It organizes reusable strategies and repair assets with validation records via the GEP protocol, offering market- and quality-based ranking to help different agents discover, evaluate, and inherit existing capabilities.
分类推荐榜首KKiro is an AI engineering tool for developers that organizes requirements into specifications, architectures, and tasks, generates code via agents, and supports the entire workflow—from planning to code review—through property-based testing, execution memory, and integration with development tools.
分类推荐榜首M
MemoAIMusic & audioMemoAI is an AI-powered audio-to-text tool that supports localized processing, eliminating the need to upload data and ensuring privacy and security. It can efficiently transcribe YouTube videos, podcasts, and local audio/video files (such as MP4, MP3, AAC, M4A, etc.) into text, while offering core features like multilingual translation, speaker diarization, real-time subtitles, and floating notes—making it ideal for meeting minutes, interview transcription, content learning, and more.
分类推荐榜首JJAX is a Python numerical computing library designed for accelerators, specifically tailored for high-performance array computations and large-scale machine learning. It efficiently executes the same code across multiple hardware backends, including CPUs, GPUs, and TPUs.
分类推荐榜首OOpenDesign is an open-source design workspace that can connect to local programming agents, use built-in model keys, or leverage cloud-based models, enabling the creation and editing of prototypes, web pages, presentations, and graphics within a design system; local software is free, while model services incur fees.
分类推荐榜首A
AI ShortWriting & contentAIS ALERT MANAGEMENT AND SHARED PLATFORM, QUICK COMMAND FOR MULTIPLE SCENARIOS
分类推荐榜首妙THE AI LIVE TELECOM SOLUTION FROM THE AD
待评分 · 分类列表推荐闪FlashCut is an AI-powered creation platform for voice-over and short-video marketing. It can generate finished videos with voiceovers, subtitles, and visuals from text, offering digital human customization, photo-to-digital-human conversion, live-stream clipping, and intelligent editing. It also provides APIs and branded mini-program solutions for businesses.
待评分 · 分类列表推荐AAssemblyAI is a speech AI platform for developers, offering pre-recorded, real-time, and short-form synchronous transcription APIs. It enables the building of voice-based applications by integrating capabilities such as speaker identification, emotion analysis, chapter and summary extraction, as well as speech agent functionality, dictation, and sensitive information processing.
待评分 · 分类列表推荐FCLIENT-ORIENTED AI SMART BODY。
待评分 · 分类列表推荐Article URL: https://trius.canton.graphics/blog/000003.html Comments URL: https://news.ycombinator.com/item?id=50497748 Points: 1 # Comments: 0
Article URL: https://github.com/aperiodic-io/tparquet Comments URL: https://news.ycombinator.com/item?id=50497648 Points: 1 # Comments: 1
Article URL: https://github.com/ZigaoWang/zigzag Comments URL: https://news.ycombinator.com/item?id=50496787 Points: 1 # Comments: 0
Article URL: https://www.shotoku.dev/ Comments URL: https://news.ycombinator.com/item?id=50496705 Points: 1 # Comments: 0
Article URL: https://github.com/SamerKolasevic29/SeharaCloud Comments URL: https://news.ycombinator.com/item?id=50494677 Points: 1 # Comments: 1
Article URL: https://github.com/DennySORA/vpn-manager Comments URL: https://news.ycombinator.com/item?id=50494471 Points: 1 # Comments: 0
Article URL: https://github.com/KevinOBytes/remoteku Comments URL: https://news.ycombinator.com/item?id=50492532 Points: 1 # Comments: 0
Memro is go bases persistent memory system that lets ai agents share context acssion session using mcp, built on top of postgreSQL Comments URL: https://news.ycombinator.com/item?id=50492093 Points: 2 # Comments: 1
I've resuscitated a project I was working on years ago.I was tired of having nested `flat_map` making my code basically unreadable just to fetch a simple value. The idea is to make simple & readable the traversing of deeply nested structures like JSON responses, while staying in Ruby & keeping its flexibility. Here it goes: graphql_response.then(&'[data][repository][issues][edges]*([node][title])') # => ["Support negative indexes", "Document aggregates"] github_prs_response.then(&'*[draft].any?') # => true It's different from jq or JSONPath in that: - it's a Ruby block so you can use it with `map`, `find`, etc. - you can call Ruby method from it - it doesn't automatically flatten arrays, you decide - it raises on `nil` by default (unless you use `&`, i.e. `[customer][address]&[city]`)Obviously, it has drawbacks: - it's yet another syntax to learn - it monkey-patches `String` which might be a blocker for someI'm really looking for feedback here, what do you think? Is it worth releasing a gem with this value proposition? Comments URL: https://news.ycombinator.com/item?id=50491929 Points: 2 # Comments: 0
We will deprecate the following models across all GitHub Copilot experiences (including Copilot Chat, inline edits, ask and agent modes, and code completions) on December 14, 2026: Model Deprecation date… The post Upcoming deprecation of selected GitHub Copilot models in December appeared first on The GitHub Blog.
Vercel Token Pass now offers $150 in monthly tokens (up from $80) for $30 per team per month, applicable to featured open-weight models via AI Gateway such as Kimi K3, Kimi K3.1 Pro, GLM 5.3, GLM 5.3 Flash, and DeepSeek V4.1 Flash. Credits reset each billing cycle with no rollover; once exhausted, requests using standard keys fall back to the team's AI Gateway balance, while Token Pass-scoped keys cease without overage charges. Available as an Early Access add-on for Pro teams with limited availability and a waitlist. Connect via the Vercel CLI command 'vercel token-pass' or use a Token Pass API key if supported. Subscribe through the AI Gateway dashboard.
XYBench, a benchmark of 8,115 real-world questions spanning programming and everyday tasks, tests AI assistants' ability to recognize and respond to users' misconceptions—such as asking how to thicken frosting when the real issue is warm cupcakes. Developed with researchers from Carnegie Mellon, NVIDIA, and Johns Hopkins, XYBench measures whether models identify underlying goals (X), address them effectively, and steer users away from flawed approaches (Y). Evaluations of models like Claude Sonnet 4.6, GPT-5, Gemini 3.1 Pro, Qwen3.5-9B, and DeepSeek-V4 Pro showed they frequently focused on the stated request rather than the actual goal, with limited success in misconception identification. Human reference answers outperformed all models, highlighting a gap in pragmatic reasoning. The benchmark uses automated and human evaluation to assess pragmatic solution, emphasis, and misconception identification, offering a tool for researchers to improve AI assistants’ ability to provide helpful advice even when users ask the wrong question.
arXiv:2611.14661v1 Announce Type: new Abstract: Kalai et al. propose stating the scoring rule in the prompt with an explicit threshold t, so that a model should answer only when its probability of being correct exceeds t. We probe each requirement of compliance on shared items, in 12 checkpoints from three post-training stages of four fully open 3-8B recipes, with exploratory models up to 32B. At t = 0.75, a line limiting the reply to an answer or "I abstain" (the abstention format) raised the mean abstention rate from 20% to 91%. Yet within this format, raising t from 0.5 to 0.9 changed it by only -0.8 percentage points. Without the line, the higher threshold even lowered abstention, and up to 40% of a checkpoint's items received decisions that no compliant policy with threshold-independent beliefs could make. Given the probability that their answer was correct, the checkpoints barely responded to the threshold in a forced-choice probe. Comparisons with self-assessment signals such as P(True) depended on the grader and on answer drift between prompts. An external threshold on recalibrated P(True) made coverage non-increasing in t by construction. Calibrated on the evaluation distribution (post hoc), it beat the abstention format at t = 0.5; calibrated on TriviaQA and applied to PopQA, however, it fell below that format at 0.75. On OLMo 3 7B, decisions trained on threshold-conditioned targets did not follow unseen thresholds consistently. A post-hoc variant trained at four thresholds answered less as t rose but still failed a pre-specified test, mainly on agreement with its training rule. A high abstention rate is not evidence of compliance: evaluations using stated scoring rules should report how often models abstain, whether decisions respond to the stated threshold, and which items they answer.
arXiv:2611.14594v1 Announce Type: new Abstract: Adversarial Machine Learning research is currently centered on the search for the epsilon-ball: minimal perturbations designed to fool models while remaining imperceptible to humans. We argue that this fixation on $\ell_p$ norms is a metrological trap that communicates a distorted perception of risk to defenders, one that is unaligned with real-world system vulnerabilities. This paradigm misleads defenders into optimizing for narrow, mathematical robustness at the expense of systemic security, while simultaneously incentivizing attackers to exploit non-overlapping alternative adversarial pathways. Ultimately, we argue that the dominance of the $\ell_p$ ball does not just fail to reduce harm---it may actively facilitate it by communicating a false sense of security against the wrong threats.
arXiv:2611.14774v1 Announce Type: new Abstract: Full-duplex spoken dialogue, in which a system listens while it speaks and can backchannel, yield the floor, and be interrupted, is usually obtained by training on hundreds of thousands to millions of hours of speech. This is costly, rewrites the language model that supplies the system's knowledge, and still leaves turn-taking either implicit or decided coarsely rather than word by word. We show that a frozen, general-purpose multimodal LLM can instead be adapted into a word-level full-duplex system with little data. Our system is a thinker-talker cascade on an 80 ms frame clock. The thinker keeps the LLM frozen and trains only lightweight adapters and a few new token embeddings (97M parameters); at every frame it emits one token of agent text, with pauses, yields, and onsets represented as tokens of the same vocabulary and kept legal by a finite-state machine. A four-stage curriculum, each stage anchored to the model's own behavior under the previous format, performs the conversion in about 30 GPU-hours. The talker reads only the thinker's discrete tokens, so the two halves are trained independently. On Full-Duplex-Bench and InteractionBench, the system is competitive with open-source full-duplex systems built through large-scale speech training: among them, it is the least likely to take the floor when the user merely pauses, backchannels most often, and leads in spoken-task accuracy, groundedness, and coherence over long conversations.
arXiv:2611.16652v1 Announce Type: new Abstract: We introduce CABOT, a computational model for autonomous mobile robots in which the discrete plane the robots operate on is not a passive stage but an active computing medium. In the classical Look-Compute-Move (LCM) model every increase in computational capability is obtained by strengthening the robots themselves, with lights, memory or wider visibility, at a cost paid once per robot. CABOT strengthens the environment instead: each cell of the grid holds a state from a finite set Q, the states evolve synchronously under a uniform local rule that also senses whether a robot stands on the cell, that is, the floor is a cellular automaton, and the Look operation returns the states of the visible cells in addition to the positions of the visible robots. The robots are as weak as we can make them: oblivious, silent, lightless, myopic with visibility range 1, without multiplicity detection and unable to write to the floor. The number of field states |Q|, rather than any robot capability, becomes the measure of the resources required. On an inert floor (|Q| = 1) such robots can neither form a line, nor gather from scattered positions, nor explore a grid. We show that a handful of states changes this. Line Formation from connected configurations is solvable with four states, and field-silently, so that the floor also reaches a fixed point, with five. Gathering from arbitrary, possibly disconnected, configurations is solvable with three states and Line Formation from arbitrary configurations with six; we prove that neither can be solved field-silently with any number of states. A single robot explores every finite grid forever with four states and every torus with two, which is optimal. Further rule sets, validated by simulation only, equalize the spacing of robots on a row, fill a pattern painted on the floor and guide robots through a maze, and indicate the range of the model.
Human motion offers roboticists a great starting point for teaching humanoid robots how to move. A human demonstration captures the coordination and timing...
The healthcare revenue cycle depends on teams investigating denied claims, checking payer rules and deciding which problems need a human’s attention. AI agents could take on some of that work, but the harder question is how to fit them into the way health systems operate. For Mount Sinai Health System Inc., that starts with clear […] The post 3 ways AI is changing the healthcare revenue cycle appeared first on SiliconANGLE.
At Block Builderlab Launch Event 2026, AI agents moved out of single-player chat windows and into shared workspaces. There they work beside people as named, auditable members of the team. Block Inc. built these tools to run itself as an intelligence company, and it is now offering them to other businesses, building on Buzz, its […] The post 4 insights from Block Builderlab Launch Event 2026: AI agents join the team as working colleagues appeared first on SiliconANGLE.
Article URL: https://blog.cip.org/p/ai-models-know-what-women-peace-and Comments URL: https://news.ycombinator.com/item?id=50470975 Points: 20 # Comments: 21
Give your curiosity direction. Make every day of reading count.