Hugging Face Daily Papers6565In 18 languages, visual languages, and full-modal models, the retrieval head remains sparse in the newly added modalities (such as audio-visual) and has a high degree of cross-modal sharing. Full-modal models mainly retain the backbone model's retrieval heads; removing these retrieval heads eliminates almost all audio-visual retrieval capabilities.
The SoMark team has open-sourced the 1.3B document parsing model OXR, ranking first on three benchmarks: OmniDocBench (97.18), olmOCR-Bench (87.84), and DocParsingBench (87.76).
Why it matters: 1.3B model trained with RSI ranks first in three document parsing rankings, and provides engineering trade-offs between plug-in architecture and free API.
Hugging Face Daily Papers6565A-RSI refreshes training tasks based on learning signal decline and uses independently verifiable programs to generate answers, enhancing benchmark test performance without external labels. The model monitors reward changes and accuracy in training logs to refresh the task pool and utilizes a static model self-programming library to generate multimodal questions and their reference answers, while automatically verifying the properties of reward and refresh rules and auditing the generated answers. In eleven benchmark tests, A-RSI increased average performance from 57.77 to 61.05 with 800 new tasks.
Baidu has open-sourced the Wenxin Unlimited OCR 2 model, which supports parsing documents totaling over 3,600 pages at once, natively supporting layout element segmentation recognition and cross-page text merging, table recognition and merging. It scored 84.6 on the MPDocBench benchmark test and 77.1 on the MarathonDocBench benchmark test.
MarkTechPost5858Edge0 releases the open-source text-to-speech model Audio8-TTS-Realistic, with 0.6B parameters supporting zero-shot voice cloning for 23 languages, achieving an average error rate of 4.86% in the CV3 benchmark test, surpassing Qwen3-TTS-1.7B and VoxCPM2, but lower than the 4B-level Higgs TTS 3. The model employs a dual autoregressive architecture, including a slow branch with 24 layers and width 896, and a fast parallel prediction branch with 4 layers, trained using a learned realism critic.
Hugging Face Daily Papers6565Research formalizes benchmark evaluation through difficulty measures to identify measurement mismatches and coverage gaps, proposing diagnostic metrics and rewrite algorithms to improve benchmark coverage; analyzing 25 popular benchmarks across 7 fields reveals that most implicitly overestimate specific difficulty areas or fail to cover most of the difficulty axis, limiting model improvement monitoring capabilities; introduces BenchFill algorithm to effectively reduce local coverage gaps.
When AI models process different languages, token consumption can vary up to 17 times, with English and other mainstream language users incurring lower costs. The implicit 'tokenizer tax' results in users of less common languages paying higher fees, affecting response speed and context capacity. AI labs need to publish evaluation results for each language and improve tokenizer design to narrow the gap.
Databricks Blog4747Databricks announces the general availability of Spark Declarative Pipelines Real-Time Mode, which achieves millisecond-level end-to-end latency through continuous streaming, parallel query stages, and instant data delivery.
Cohere Blog3434Use Cohere Embed, Rerank, and Parse combined with Amazon SageMaker AI, OpenSearch Serverless, and Amazon Bedrock to search financial reports and generate answers with page-level citations.
MarkTechPost4747Prime Intellect released Prime Mega MoE, adding fast overlapping reverse transmission on the basis of DeepSeek's integrated Mega MoE core, making it available for training.
Epoch AI6565Epoch AI analyzes global AI chip sales data, measuring computing capacity in H100 equivalent units, showing exponential growth each quarter, indicating computing capacity doubles every 9 months.
NVIDIA Technical Blog3535NVIDIA significantly accelerates machine learning atomic potential energy calculations by optimizing Tensor Core performance. Its technology can improve the computational efficiency of quantum mechanics methods such as Density Functional Theory (DFT), balancing cost and accuracy. This solution is suitable for large systems or long-time simulations, but the specific parameter scale, benchmark scores, and availability (open source/closed source/API) were not provided in the original text.
Hugging Face BlogProject ProtoStar is announced, aiming to unlock the potential of publicly funded scientific research by making datasets, models, and code more accessible and reusable. The project brings together universities, research institutions, industry, and the open science community, supporting the release of scientific achievements in AI-compatible, machine-actionable formats, and demonstrating its value through a series of public challenges. The first challenge, hosted by Carnegie Mellon University, will launch on November 2.
Hugging Face BlogHugging Face releases Granite Switch 4.1.1, an incremental update based on 4.1, maintaining full backward compatibility. New features include multilingual speech-to-text models (supporting 10 languages with native punctuation) and a decision adapter for improved performance on the Jev decision index test set. The team achieves rapid deployment by updating only the adapter without recalculating the base model, and future support for third-party custom adapters is planned.
Today is auspicious for rest at ISHO Today is auspicious for rest AI Sleep Technology Ecosystem Product Launch Conference announced its strategy, focusing on precise monitoring, scientific intervention, and energy recovery, planning to collect over 1000 cases of gold standard clinical PSG data within a year, and extending the research scope from nighttime sleep quality to daytime energy status.
DAYJOB: Tax evaluates the performance of AI agents in 50 professional tasks across 8 tax areas, requiring agents to complete tax filing, compliance, and advisory work. Claude Sonnet 5.5 leads the rankings with a pass rate of 10.8%, while Claude Opus 5.5 has a pass rate of 9.2%. The benchmark includes 1,789 workspace files, covering various tax scenarios and complex materials, created and reviewed by tax experts.
MolCell AI's MMDesign platform completes 18 design projects with a success rate of 94.7%, hitting all six GPCR targets among them, with the best affinity reaching nanomolar level; the platform uses a 'generate-screen-verify' workflow, generating tens of thousands of candidate molecules in computational space and testing only 9-50 to achieve success, with an overall candidate hit rate of 30.7%, demonstrating efficient and precise molecular design capabilities.
Hugging Face Daily PapersRSI Arena demonstrates that autonomous AI intelligence can independently use human feedback to train and iteratively improve language models. Most agents achieved preference gains in the second phase, although benchmark test improvements were limited. Ten cutting-edge models were used as training agents, starting from NVIDIA Nemotron 3.5 Lightning 30B-A3B Base, which independently selected data, implemented training, and selected checkpoints in the first phase within six days, using a 64-GPU cluster and 300 USD API credits.
OpenRouter BlogThe Anthropic/claude-haiku-5.5 model paired with a strict JSON schema can extract tables from PDFs and return typed rows. Tested five PDFs (including scans) successfully extracted all 644 cells; failure cases were due to PDF processing paths, not model capability. Note empty results and output length limits.
Databricks BlogProcore uses OpenSharing to programmatically share connected construction data into clients' existing environments, supporting Power BI, Tableau, Snowflake, BigQuery, Databricks workspaces or local SQL Server for reporting and downstream analytics.
Anthropic NewsroomAnthropic expands the AI for Science project, providing free Claude credits to 20 global research institutions, foundations, and research centers to support high-impact science research.
LangChain BlogOpenWiki makes agents about 5 times more accurate on cross-source connection issues than single search, adding evidence images increases accuracy by another 27% and doubles citation consistency.
TechCrunch · AINVIDIA has released the open-source speech-to-text model Nemotron 3.5 Transcribe, supporting transcription in 40 languages, containing 3 billion parameters, and running on Jetson AGX Thor and DGX Spark chips. The model can translate 31 languages, distinguish 8 speakers, and surpass competitors in the Hugging Face Open ASR Leaderboard benchmark tests, but only supports recording audio.
MooreThread open-sources Megatron-MUSA-Titan project, supporting trillion-parameter MoE model training for multi-thousand node clusters, covering pre-training and reinforcement learning scenarios, achieving high MFU and long-stable training. The project supports blockwise FP8 training, combining DeepEP and ACE asynchronous communication engines to accelerate token distribution and result merging, improving end-to-end training efficiency.
Survival LLM is a free AI assistant and emergency knowledge base that runs entirely on a personal computer, providing answers based on trusted sources without the internet and displaying the answer sources.
MarkTechPost6060FlashMap replaces the discrete diffusion drawer with a continuous flow drawer, achieving up to 6.14 times lossless speed improvement on Qwen3-8B for GSM8K. Its drawer weights (Qwen3-8B and Qwen3-4B) have been open-sourced to Hugging Face, supporting vLLM and SGLang plugins.
Developers found three model identifiers—deepseek-v5, deepseek-v5-open, and deepseek-v5-next—in the DeepSeek Harness GitHub repository, currently serving as framework placeholders.
Hugging Face Daily Papers6565Large-scale evaluations show that dedicated memory systems are often weaker than simple retrieval-enhanced generation. The Memory Benefit and Memory Tax framework is introduced to evaluate the net value. Memory is becoming increasingly common in long-term intelligent agents, but the net value of building memory compared to directly retrieving from raw history remains unclear.
Hugging Face Daily Papers6565HABIT-Bench evaluates an agent's ability to learn habits from synthetic user histories involving hundreds of interactions, requiring the agent to integrate scattered evidence and choose appropriate actions for new requests. Using Qwen3-235B-A22B and GLM-5.3-Flash, eight memory methods were evaluated, with the best configuration achieving a macro-average accuracy of 33.61% in a four-choice task, outperforming random guessing at 25%. Even with labeled evidence, there were still numerous errors, indicating that understanding how to guide decisions is more important than simply recalling history.
Hugging Face Daily PapersHarness Self-Evolution improves LLM agents by guiding skill adaptation without parameter updates, enhancing the performance of weak models. The study found that skills are beneficial for strong models but potentially harmful for weak models. In seven models and three benchmark tests, 103 task-skill pairs improved GPT-5.6-luna by 31.1 points but reduced Gemma-4-E4B by 8.8 points.
Hugging Face Daily PapersOvisOCR3 is a 0.8B visual language model that unifies multiple OCR tasks through multi-mentor distillation and reinforcement learning, achieving SOTA results on document parsing, general OCR, key information extraction, image-text description, and cross-page table reconstruction tasks. It scored 96.74 on OmniDocBench v1.6, ranked first in OCRBench v2 English and Chinese split, and surpassed baseline models in multiple benchmark tests.
Hugging Face Daily Papers6565Redistribution sensitivity score (RSS) identification generates repeatable feature redistribution patterns on 27 pre-trained SAE models (covering 8 language models), combined with clean margin to quantify its contribution to downstream reading.
Ars Technica · AI6565Pinal can generate proteins through natural language prompts, with a success rate meeting specific functional requirements; Hugging Face also launches a similar tool at the DNA level. Pinal uses T2struct to convert prompts into structural blueprints, which are then combined with SAProt-T for sequence generation based on probabilistic evaluation, trained based on three major databases and AlphaFold predictions.
Databricks BlogDatabricks Lakehouse Federation allows users to query data directly from external databases (such as SQL Server, PostgreSQL, etc.) using standard SQL without migration.
Databricks Blog4545Databricks Vital Skills enhances Genie Code's knowledge in the healthcare and life sciences field through the open-source Agent Skills standard, achieving a score improvement in the BiomniBench benchmark test.
AWS Machine Learning Blog3434Property Finder discovered that 96.8% of the labels in its evaluation dataset for visual AI workloads migrated to Amazon Bedrock were predictions from a single model rather than human validation, hindering objective comparison of new models. The company built a trustworthy benchmark dataset by independently labeling 356 images with human experts and, after 27 iterative optimizations with evolutionary prompt tuning, reduced costs by 86% while maintaining an 86% accuracy rate.
Hugging Face BlogHugging Face releases Old Ideas, Novel Problems study, evaluating the reliability of six LLMs (GPT-5.1 to 5.4, Claude Sonnet 4.5 to 4.6) in judging the reliability of research innovation.
