保存済みツールを推薦
FOXAI SEARCH

サイト内検索

ニュース記事 220

モデル更新

Same Rule, Different Verdicts: Rubric Language and Clause Structure in LLM Judges for Korean Support Dialogues

arXiv:2611.04091v1 Announce Type: new Abstract: Services that operate in Korean often write their evaluation rubrics in Korean and hand them to an LLM judge. We test whether a judge's verdict depends on the language of the rubric and on how a rubric item is phrased. We write 12 criteria for support dialogues in four clause structures (affirmative, negated, conditional, exception) such that two pairs of structures have the same meaning, and we give each item in Korean and in English. Six low-cost models from three providers, plus one reference model, judge 60 translated ESConv dialogues and 144 constructed boundary cases under every combination (51,744 requests across the seven models). The analysis plan was fixed and committed before data collection. Cross-provider agreement is lower with Korean rubrics than with English rubrics (Cohen's kappa 0.687 vs. 0.745; difference -0.058, 95% CI [-0.072,-0.045]). Disagreement concentrates in negated, conditional, and exception items (odds ratio 5.6 relative to affirmative items). A single model changes its verdict between two items with the same meaning in 11-15% of cases. Two hypotheses are not supported: models from the same provider do not agree more than models from different providers, and adding paired examples does not reduce the language gap. We measure consistency only; no human labels are used, and all materials were drafted by an LLM.

翻訳サービスが停止中のため原文を表示しています。
arXiv cs.CL
model-evals

Say Less, Mean Better: Training-Free Verbosity Control for Language Models

arXiv:2611.04107v1 Announce Type: new Abstract: Large language models (LLMs) often generate responses longer than necessary, increasing inference latency and computational cost. While prompting can encourage conciseness, controlling output length without compromising response completeness remains challenging. Aggressive early stopping can reduce generation costs but risks terminating responses prematurely. We introduce MinimalLM, a training-free framework for reliable verbosity control that combines uncertainty-aware, length-dependent stop-token adjustment with structural completeness guardrails. MinimalLM separates the decision of when to stop from whether stopping is structurally appropriate, coordinating response-level instructions, decoding-time intervention, and completion checks without modifying model parameters or requiring auxiliary models. We evaluate MinimalLM across three open-source LLMs and 330 questions, producing 5,940 responses across baseline and ablation configurations. MinimalLM reduces generated tokens by 47.6%, accelerates individual responses by 19-68%, and eliminates structurally truncated outputs in the primary evaluation. Despite substantial token savings, it achieves an average quality score of 8.04/10, compared with 8.42/10 for unmodified models and 6.46/10 for length-matched truncation. Evaluations across diverse subject domains (such as healthcare, programming, law, and science) demonstrate robust efficiency gains, while ablations establish the complementary roles of decoding control and completeness constraints. Our findings highlight that efficient generation requires more than producing fewer tokens: it requires controlling when generation ends while preserving response integrity. MinimalLM offers a lightweight, training-free approach toward this goal.

翻訳サービスが停止中のため原文を表示しています。
arXiv cs.CL
ai-safety

Securing the agentic enterprise with data lineage

A sales representative asks an AI agent for customer pricing, and the agent pulls data from multiple sources: restricted pricing systems, internal communications, and archived quotes. The output looks accurate to the agent, so a second agent uses the data to draft the proposal for the sales rep. No human checks either handoff. When it’s…

翻訳サービスが停止中のため原文を表示しています。
MIT Technology Review · AI
startup-projects

Show HN: Alternative to Nvidia Auto-Ontology

This repo provides an alternative to the ontology compiling capabilities of NVIDA Auto-Ontology.The main difference is that it builds the ontology from most of your existing data assets (transformation code, dashboards, docs), instead of building it from your warehouse and semantic layer only.It consists in skills and python scripts that ingest the metadata and docs your team already has, and compile them into YAML and Markdown files structured in a file tree that an agent can use as context to query your data.When your existing asset contradict, or reveal a gap in the business context around your data, it writes down questions for you and your team instead of guessing a plausible meaning.It runs locally in any coding agent. Comments URL: https://news.ycombinator.com/item?id=50363667 Points: 1 # Comments: 0

翻訳サービスが停止中のため原文を表示しています。
Hacker News Show HN
モデル更新

Swapping the Judge Changes the Ranking: A Preregistered Study of Cross-Provider Replication in LLM-Judged System Comparisons

arXiv:2611.04090v1 Announce Type: new Abstract: Practitioners who compare systems with an LLM judge and have no human labels often check a conclusion by repeating it with a judge from another provider. We test how far that check can be trusted. In a preregistered study (with a disclosed post-registration reduction of the judge set), five judge models from three providers re-judged 4,000 Chatbot Arena battles (nine systems, 33 system pairs) that carry human votes, each in both response orders. Agreement with the human vote was low at the verdict level (median Cohen's kappa = 0.22), and system rankings did not survive a change of judge: Kendall's tau with the human ranking ranged from 0.17 to 0.89 (median 0.67), so the preregistered hypothesis that rankings remain stable was not supported. Conclusions replicated by two low-cost judges from different providers matched the human conclusion more often than single-judge conclusions (+0.10 precision), but the gain came from filtering out one poorly aligned judge, and replication involving that judge left a conclusion for only 35% of comparisons. Judges that agreed with humans reached a stable conclusion with no more verdicts than human votes required; the others failed to converge on 26-70% of system pairs at the sample sizes tested (up to 100 verdicts per pair). Disagreements with the human vote co-occurred across judges far more often than independence predicts, although a single human vote per battle cannot separate shared judge bias from voter noise.

翻訳サービスが停止中のため原文を表示しています。
arXiv cs.CL
model-evals

Towards Automatic Heart Murmur Characterization from Multi-spot Phonocardiograms

arXiv:2611.03654v1 Announce Type: new Abstract: Cardiac auscultation traditionally detects the presence of a heart murmur, but clinicians further characterize it by timing, shape, pitch, quality, and grade. Most automatic phonocardiogram analysis systems stop at binary detection and do not estimate this richer description. In this work, we formulate automatic murmur characterization as five patient-level classification tasks and compare three multi-spot strategies on recordings from the aortic, pulmonary, mitral, and tricuspid auscultation locations. First, classical and deep-learning classifiers are trained independently at each location, and predictions are combined via a maximum-severity late-fusion rule. Second, a synchronous multi-input CNN jointly processes one representative segment per location for a single task. Third, a unified multi-input, multi-output CNN predicts all five attributes from one shared multi-spot representation. Using 735 patients from the CirCor DigiScope dataset with complete four-spot recordings, the best late-fusion configurations reach an unweighted mean sensitivity between 85.56% and 93.81% across the five tasks, with class-specific recalls between 80.00% and 100.00% for every relevant category. The multi-input CNN trails late fusion by 7-16 points depending on the task, while the multi-output model reaches values between 72.59% and 79.40%, estimating all five attributes from one shared representation. We further show that naively balancing training data by adding all minority-class segments degrades patient-level performance under the maximum-severity rule, since it introduces non-representative, noisy segments not filtered by the clustering-based segment selection used for evaluation. These results indicate multi-spot murmur characterization is feasible with both fusion-based and jointly learned architectures, outlining a path toward automatic, structured murmur reports complementing binary screening.

翻訳サービスが停止中のため原文を表示しています。
arXiv cs.LG
agent-frameworks

v1.19.0 - Trace Your Workflows, Layer by Layer

Dify 1.19.0 enhances observability with nested workflow tracing, allowing users to follow execution steps inside Workflow Tools, inspect LLM calls, and maintain context across nested invocations. The release also enables portable app packages via App packages, improves Agent file uploads and previews, introduces Skill Hub automation through CLI commands, and upgrades the bundled plugin daemon to version 0.6.14-local. Key changes include support for format v2 app packages, enhanced security, improved editing workflows, and updated environment variables and API endpoints.

翻訳サービスが停止中のため原文を表示しています。
Dify releases
coding-engineering

Vercel Security Review now in public beta

Vercel Security Review is now available in public beta for Pro and Enterprise teams, automatically scanning code pushes to connected GitHub repositories for vulnerabilities. Reviews trigger on every push, including direct and force pushes, and findings are reported in a dashboard with details like project, environment, branch, commit, summary, suggested fix, and AI-generated severity. Findings appear as GitHub checks or review comments, and users can fix issues via Vercel Agent or CLI. Teams can set security policies for context, but the tool does not perform runtime or penetration testing. Pricing follows Vercel Agent rates, with free access until November 10, 2026.

翻訳サービスが停止中のため原文を表示しています。
Vercel