Tools & Code

Developer tools, open-source releases, frameworks, IDE integrations, dev workflows.

TikStance dataset enables multimodal stance detection in short-form political video

Researchers have released TikStance, a multimodal dataset linking 161 political videos with 13,876 hierarchical comments from the 2024 U.S. election cycle. The resource preserves both audiovisual and conversational context for stance detection, addressing a critical gap in training data for short-form video analysis. As political discourse migrates to platforms like TikTok, this dataset enables development of models that understand nuanced positions across modalities and nested discussions, directly supporting the next generation of content moderation and political discourse analysis systems.

arXiv cs.CL·2d ago

Research Tools & Code

Researchers enable tokenizer upgrades for deployed language models

Tokenizer vocabulary is typically locked at pre-training time, creating a structural problem when deployment priorities shift: languages added later fragment into many tokens per word, inflating latency and compute costs for those users. On-device models face particular pressure since embedding and output matrices consume substantial decode bandwidth. This paper introduces an in-place tokenizer expansion technique that lets model producers upgrade vocabulary post-hoc without full retraining, addressing a real efficiency gap between cloud and edge deployments. The work targets a concrete pain point in multilingual and evolving model ecosystems.

arXiv cs.LG·2d ago

Products & Apps Tools & Code

Google bundles compute into Gemini Notebook, opens Search to third-party apps

Google is consolidating its AI notebook product under the Gemini brand while expanding its infrastructure capabilities. Each Gemini Notebook now includes dedicated cloud compute for code execution, initially available to AI Ultra and Workspace subscribers. This move signals Google's strategy to deepen Gemini integration across its product suite and compete with Jupyter-adjacent tools in the AI development workflow. Simultaneously, Google Search gains third-party app connectors, positioning search as an extensible platform rather than a closed service. Together, these changes reflect Google's broader push to make Gemini the connective tissue across consumer and enterprise AI experiences.

The Decoder·2d ago

Business & Funding Tools & Code

Enterprise AI agents outpace the data governance needed to trust them

Enterprise AI deployments are hitting a critical inflection point: retrieval-augmented generation has become standard practice, yet a majority of organizations report their agents confidently producing incorrect answers due to inconsistent or missing business context. The shift from dedicated vector databases to provider-native retrieval tools is accelerating, but trust in the underlying data layer lags behind infrastructure speed. A governed semantic layer is emerging as the industry's answer, though most enterprises are still in early implementation. This context gap represents a fundamental architectural challenge that will shape how enterprises architect AI systems over the next 18 months.

VentureBeat - AI·2d ago

Research Tools & Code

Clinicians build safety taxonomy for medical AI model failures

Medical AI safety has lacked systematic failure taxonomy. MedFailBench introduces a clinician-authored benchmark that categorizes model errors by severity and failure mode, not just accuracy. The framework identifies six distinct safety gates: missed escalations, unsafe dosing, inappropriate discharge reassurance, hallucinated evidence, protocol violations, and unsupported claims. This shifts evaluation from binary correctness toward granular risk profiling, enabling developers to stress-test models against realistic clinical failure patterns. The open-source release with automated screening pipelines establishes infrastructure for safety-focused model iteration in healthcare, addressing a gap where traditional benchmarks miss high-stakes boundary violations.

arXiv cs.CL·2d ago

Products & Apps Tools & Code

DoorDash launches command-line tool designed for AI agents

DoorDash's command-line interface represents a deliberate shift in how consumer platforms architect for non-human users. By exposing ordering workflows through terminal-native tooling, the company signals that AI agents are now a primary design consideration alongside traditional app interfaces. This move reflects a broader infrastructure transition where companies must support both human UX and machine-readable APIs as first-class concerns. For developers and AI teams, it lowers friction for autonomous ordering workflows and sets a precedent for how legacy consumer services adapt to agent-native architectures.

TechCrunch - AI·2d ago

Products & Apps Tools & Code

OpenAI frames Codex as builder velocity multiplier for France's developer community

OpenAI's Romain Huet highlights how Codex is enabling developers to prototype and ship ambitious projects faster, framing code generation as a force multiplier for builder velocity. The framing matters: positioning AI-assisted development as a cultural shift toward experimentation rather than a tool feature suggests OpenAI sees Codex adoption as tied to developer confidence and regional tech ecosystems. For practitioners, this signals OpenAI's continued investment in lowering friction between ideation and execution, a key lever for expanding the developer base beyond early adopters.

OpenAI (YouTube)·2d ago

Models & Releases Tools & Code

Thinking Machines Lab releases 975B open-weights model Inkling

Thinking Machines Lab, led by Mira Murati, has released Inkling, a 975B-parameter mixture-of-experts model under Apache 2.0 licensing. The multimodal system trained on 45 trillion tokens represents a significant open-weights entry from a new lab, challenging the concentration of frontier model releases among established players. A smaller 276B variant is forthcoming, signaling a tiered release strategy. The notably sparse model card raises questions about documentation standards in the open-weights ecosystem, even as the release itself expands accessible frontier-scale infrastructure.

Simon Willison·2d ago

Research Tools & Code

AlphaWiSE interpolates multimodal checkpoints to balance continual learning tradeoffs

Continual learning in multimodal systems faces a fundamental tension: adapting to new data often erodes the cross-modal alignment learned earlier. AlphaWiSE addresses this by interpolating between frozen checkpoints in weight space, allowing practitioners to dial the stability-plasticity tradeoff per parameter rather than committing globally. The method fits scalar coefficients on a small exemplar buffer, producing a single deployable model without architectural overhead. This matters for production CLIP-like systems that must absorb streaming data without catastrophic forgetting or expensive retraining cycles.

arXiv cs.LG·2d ago

Research Tools & Code

Self-validating rubrics emerge from queries without human labels

Researchers propose Rubrics on Trial, a method for automatically generating and validating evaluation rubrics from user queries alone, without human annotation or model retraining. The framework bootstraps rubric quality by synthesizing response pairs conditioned on candidate rubrics, then tests each proposal's ability to meaningfully distinguish answer quality before incorporation. This addresses a critical bottleneck in LLM training and evaluation: the difficulty of constructing reliable, task-specific scoring criteria. For practitioners building custom evaluators or fine-tuning models, this reduces dependency on expensive human-labeled preference data while maintaining rigor through synthetic validation.

arXiv cs.CL·2d ago

Tools & Code Products & Apps

Willison uses Claude to port Mermaid ASCII converter to WebAssembly

Simon Willison has expanded his Mermaid diagram conversion toolkit by compiling an older Go library into WebAssembly using Claude Fable 5, enabling browser-based rendering with color support. This represents a practical workflow for converting LLM-friendly diagram syntax into terminal-compatible ASCII art, addressing a gap between modern diagramming tools and legacy systems. The move signals how AI-assisted compilation is enabling developers to resurrect and adapt older codebases for contemporary web environments, particularly useful for teams bridging visual design and infrastructure automation.

Simon Willison·2d ago

Research Tools & Code

Sparse equation recovery method scales to practical engineering problems

SINDy represents a meaningful counterweight to the data-hungry neural network paradigm dominating surrogate modeling in engineering. By recovering sparse, interpretable equations from small datasets through regression over nonlinear term libraries, the method addresses a persistent friction point: practitioners often lack the massive labeled datasets required for deep learning, yet need models that expose their underlying physics rather than acting as black boxes. This tutorial bridges the gap between theoretical validation on toy problems and real-world deployment, making symbolic regression techniques more accessible to domain experts who prioritize explainability and sample efficiency over raw predictive power.

arXiv cs.LG·2d ago

Tools & Code Hardware & Infra

OpenAI and Work Louder launch Codex Micro joystick controller for AI agents

OpenAI and Work Louder have co-developed the Codex Micro, a hardware controller that shifts AI agent interaction from text commands to joystick-based control. This move signals a broader industry pivot toward more intuitive, real-time interfaces for autonomous systems, potentially lowering the barrier for non-technical users to supervise and steer AI agents. The hardware play reflects growing recognition that keyboard-centric workflows may constrain how developers interact with increasingly autonomous models, opening a new category at the intersection of developer tools and human-AI collaboration.

The Decoder·2d ago

Research Tools & Code

New visualization framework tackles interpretability gap in categorical machine learning

Researchers introduce cGAP, a visualization framework that addresses a persistent gap in machine learning tooling: interpretable exploration of high-dimensional categorical data. Unlike existing methods that either collapse to low-dimensional projections or sacrifice readability for predictive power, cGAP preserves the original data matrix while embedding subjects and category levels in three-dimensional space mapped to RGB coordinates. The work targets domains where categorical structure dominates (genetics, biomedicine, social science) and reflects growing recognition that interpretability infrastructure lags behind model capability, particularly for non-continuous modalities that remain common in real-world applications.

arXiv cs.LG·2d ago

Research Tools & Code

Simulation method synthesizes formally verified control policies

Formal verification remains a critical bottleneck for deploying learned controllers in safety-critical systems. SMC-ES addresses this by combining simulation-based policy synthesis with probabilistic guarantees on safety, robustness, and performance, eliminating the traditional trade-off between learning flexibility and provable correctness. This bridges reinforcement learning's scalability with the formal assurance requirements of autonomous vehicles, robotics, and industrial control, potentially unlocking deployment pathways currently blocked by certification demands.

arXiv cs.LG·2d ago

Research Tools & Code

LongStraw enables million-token RL training on fixed GPU budgets

LongStraw addresses a critical bottleneck in AI agent development: the ability to run reinforcement learning post-training on million-token contexts within fixed GPU budgets. Current RL systems plateau at 256K tokens, forcing length generalization at deployment time, which undermines agents that accumulate observations and tool outputs over extended trajectories. This architecture-aware execution stack uses Group Relative Policy Optimization to eliminate redundant autograd computation, cache only token-specific state, and replay response branches sequentially, trading compute time for memory efficiency. The work signals growing recognition that agent capability depends on training-time context depth, not just inference-time window size.

arXiv cs.LG·2d ago

Business & Funding Tools & Code

Visa, Mastercard, Google and Microsoft back x402 payments standard

A coalition of 40 major technology and financial firms, including Visa, Mastercard, Google, and Microsoft, has established a foundation to govern the x402 payments protocol. This standardization effort addresses a critical infrastructure gap as AI systems increasingly require seamless, interoperable transaction capabilities. The move signals that payment rails for AI services and agent-to-agent commerce are transitioning from proprietary implementations to industry-wide standards, similar to how HTTP unified web infrastructure. For AI builders, this reduces friction in monetizing models and services across platforms. The backing of both fintech and cloud giants suggests the protocol will likely become foundational for future AI commerce ecosystems.

AI Business·2d ago

Models & Releases Tools & Code

Google silently patches Gemma 4 tool calling and truncation bugs

Google pushed a quiet update to Gemma 4 that addresses critical production issues: tool calling failures, response truncation, and GPU performance gaps on Nvidia Hopper hardware. The stealth deployment under an unchanged version number signals Google's pragmatic approach to open-model maintenance, where fixes ship without fanfare or version bumps. For practitioners relying on Gemma 4 in production, this matters because tool use and response completeness are table-stakes for real-world LLM applications. The move also reflects competitive pressure to keep open alternatives competitive with proprietary systems on infrastructure that matters most to enterprise deployments.

The Decoder·2d ago

Research Tools & Code

Execution-verified code distillation improves financial reasoning in smaller models

Researchers have developed a distillation method that transfers numerical reasoning capabilities from large language models to smaller ones by using execution-verified Python programs as supervision signals rather than natural-language explanations. The approach addresses a critical weakness in LLM-based financial reasoning: arithmetic errors in textual rationales that corrupt training data. By filtering for programs that execute correctly and match gold answers, the technique ensures higher-quality knowledge transfer for domain-specific tasks requiring hybrid reasoning across tables and text. This matters for practitioners building compact financial AI systems that must balance accuracy with inference efficiency.

arXiv cs.CL·2d ago

Research Tools & Code

Smaller models with scaffolding beat larger ones in high-stakes supervision

Researchers demonstrate that wrapping smaller language models in deterministic scaffolding (retrieval systems, schema validation, human-in-the-loop gates, audit trails) outperforms larger unstructured models in high-stakes domains. The case study compares a baseline GPT-5 chatbot against a GPT-4o-mini system embedded in a LangGraph harness for academic supervision, revealing a critical shift in production AI: raw model scale matters less than architectural composition when reliability and accountability are non-negotiable. This work signals growing maturity in operationalizing LLMs beyond chat, with implications for regulated industries where explainability and auditability trump raw fluency.

arXiv cs.CL·2d ago

Tools & Code Policy & Regulation

xAI releases Grok-Build source code after credential-upload vulnerability

xAI's Grok-Build tool exposed a critical infrastructure vulnerability when it automatically uploaded sensitive user data, including cryptographic keys and credential stores, to Google Cloud without explicit consent. Following public backlash, the company committed to data deletion and released the entire 844,530-line Rust codebase under Apache 2.0 licensing. The incident underscores persistent security gaps in AI developer tooling and raises questions about data handling practices across the emerging AI infrastructure layer, while the open-source release may accelerate community-driven security audits.

The Decoder·2d ago

Research Tools & Code

Adaptive draft pruning improves speculative decoding under concurrent load

Speculative decoding has emerged as a key efficiency lever for LLM inference, but its gains erode under high concurrency when long draft sequences waste compute on rejected tokens. D-Cut addresses this by dynamically pruning draft tokens across batches, routing verification budget toward high-confidence predictions. The technique targets a real production bottleneck: as inference systems scale to handle multiple concurrent requests, naive speculative strategies can actually underperform standard autoregressive decoding. This work matters for anyone operating inference infrastructure at scale, particularly in cost-sensitive deployments where batch efficiency directly impacts margins.

arXiv cs.CL·2d ago

Tools & Code

Willison extracts Grok's Mermaid renderer, ports to browser WebAssembly

Simon Willison reverse-engineered a Rust-based Mermaid diagram renderer from Grok's open-sourced codebase and ported it to WebAssembly for browser use. The tool converts Mermaid syntax into Unicode box art, enabling terminal-friendly diagram rendering without external dependencies. This exemplifies how AI agent infrastructure components are being extracted, repurposed, and made accessible to developers as reusable building blocks. The move signals growing modularity in AI tooling ecosystems and highlights how open-sourced agent codebases become raw material for downstream developer tools.

Simon Willison·3d ago

Tools & Code Policy & Regulation

xAI's grok CLI exposed user data before going open source

xAI's grok CLI tool triggered a security crisis after users discovered it was uploading entire directories to Google Cloud without explicit consent or warning. The incident exposed a fundamental trust violation in developer tooling: a utility designed to streamline workflows became a vector for exfiltrating sensitive data including SSH keys, credentials, and personal files. The subsequent open-sourcing appears to be damage control, but the episode underscores how rapidly AI infrastructure tools can erode user confidence when privacy safeguards are absent or opaque. For teams evaluating xAI integrations, this signals the need for rigorous vetting of CLI tools and data handling practices.

Simon Willison·3d ago

Research Tools & Code

RAG-based system pushes LLMs toward document-level translation coherence

Researchers introduce PAT, a retrieval-augmented translation system that moves LLMs beyond sentence-level processing to handle full documents with cultural and rhetorical coherence. By anchoring generation to corpus examples spanning paragraph and document scope, the work targets a real gap in machine translation: preserving discourse patterns and pragmatic norms across language pairs where structural differences matter. This represents a meaningful shift in how practitioners might scaffold LLMs for professional translation workflows, where context fidelity directly impacts usability.

arXiv cs.CL·3d ago

Tools & Code Products & Apps

OpenAI open-sources Codex gesture control for workplace automation

OpenAI has released Codex integration tooling that transforms physical gestures into automated workplace actions, enabling users to trigger meeting exits, calendar blocks, and system logouts via a simple red card signal. This positions LLM-powered agents deeper into enterprise workflow automation, moving beyond text interfaces toward embodied control systems. The open-source release signals OpenAI's strategy to embed Codex capabilities into developer ecosystems, lowering friction for workplace automation adoption and expanding the surface area where language models mediate human-computer interaction in professional settings.

OpenAI (YouTube)·3d ago