Tootfinder

Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks
Yuxiang Zhang, Jiangming Shu, Ye Ma, Xueyuan Lin, Shangxi Wu, Jitao Sang
https://arxiv.org/abs/2510.12635

Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks
Large Language Models face challenges in long-horizon agentic tasks as their constrained memory is easily overwhelmed by distracting or irrelevant context. Existing working memory methods typically rely on external, heuristic mechanisms that are decoupled from the agent's core policy. In this work, we reframe working memory management as a learnable, intrinsic capability. We propose a novel framework, Memory-as-Action, where an agent actively manages its working memory by executing explicit edi…

Give yourself the gift of digital permanence by applying for a DPOE grant to pay for digital preservation training this spring.
Students who've applied it to my online DIG 550 class https://DigitalCuration.UMaine.edu say it's easy and paid for the course.
---

Iterative Data Curation with Theoretical Guarantees
V\"ain\"o Yrj\"an\"ainen Johan Jonasson, M{\aa}ns Magnusson
https://arxiv.org/abs/2510.11428 https://

Iterative Data Curation with Theoretical Guarantees
In recent years, more and more large data sets have become available. Data accuracy, the absence of verifiable errors in data, is crucial for these large materials to enable high-quality research, downstream applications, and model training. This results in the problem of how to curate or improve data accuracy in such large and growing data, especially when the data is too large for manual curation to be feasible. This paper presents a unified procedure for iterative and continuous improvement …

Pharmacist: Safety Alignment Data Curation for Large Language Models against Harmful Fine-tuning
Guozhi Liu, Qi Mu, Tiansheng Huang, Xinhua Wang, Li Shen, Weiwei Lin, Zhang Li
https://arxiv.org/abs/2510.10085

Pharmacist: Safety Alignment Data Curation for Large Language Models against Harmful Fine-tuning
Harmful fine-tuning issues present significant safety challenges for fine-tuning-as-a-service in large language models. Existing alignment-stage defenses, e.g., Vaccine, Repnoise, Booster, and T-Vaccine, mitigate harmful fine-tuning issues by enhancing the model's robustness during the alignment phase. While these methods have been proposed to mitigate the issue, they often overlook a critical upstream factor: the role of the original safety-alignment data. We observe that their defense perform…

from my link log —
The significance of Brooker's autocodes in taking the early Manchester machines into the market.
https://www.curation.cs.manchester.ac.uk/computer50/www.computer50.org/mark1/gethomas/manchester_autoc…

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 2724 nodes and 14322 edges.
Tags: Biological, Protein interactions, Unweighted…

SpaceVista: All-Scale Visual Spatial Reasoning from mm to km
Peiwen Sun, Shiqiang Lang, Dongming Wu, Yi Ding, Kaituo Feng, Huadai Liu, Zhen Ye, Rui Liu, Yun-Hui Liu, Jianan Wang, Xiangyu Yue
https://arxiv.org/abs/2510.09606

SpaceVista: All-Scale Visual Spatial Reasoning from mm to km
With the current surge in spatial reasoning explorations, researchers have made significant progress in understanding indoor scenes, but still struggle with diverse applications such as robotics and autonomous driving. This paper aims to advance all-scale spatial reasoning across diverse scenarios by tackling two key challenges: 1) the heavy reliance on indoor 3D scans and labor-intensive manual annotations for dataset curation; 2) the absence of effective all-scale scene modeling, which often …

ACMID: Automatic Curation of Musical Instrument Dataset for 7-Stem Music Source Separation
Ji Yu, Yang shuo, Xu Yuetonghui, Liu Mengmei, Ji Qiang, Han Zerui
https://arxiv.org/abs/2510.07840

ACMID: Automatic Curation of Musical Instrument Dataset for 7-Stem Music Source Separation
Most current music source separation (MSS) methods rely on supervised learning, limited by training data quan- tity and quality. Though web-crawling can bring abundant data, platform-level track labeling often causes metadata mismatches, impeding accurate "audio-label" pair acquisi- tion. To address this, we present ACMID: a dataset for MSS generated through web crawling of extensive raw data, fol- lowed by automatic cleaning via an instrument classifier built on a pre-trained audio encoder tha…

Domain Knowledge Infused Generative Models for Drug Discovery Synthetic Data
Bing Hu, Jong-Hoon Park, Helen Chen, Young-Rae Cho, Anita Layton
https://arxiv.org/abs/2510.09837 ht…

Domain Knowledge Infused Generative Models for Drug Discovery Synthetic Data
The role of Artificial Intelligence (AI) is growing in every stage of drug development. Nevertheless, a major challenge in drug discovery AI remains: Drug pharmacokinetic (PK) and Drug-Target Interaction (DTI) datasets collected in different studies often exhibit limited overlap, creating data overlap sparsity. Thus, data curation becomes difficult, negatively impacting downstream research investigations in high-throughput screening, polypharmacy, and drug combination. We propose xImagand-DKI, …

Convergence and Stability Analysis of Self-Consuming Generative Models with Heterogeneous Human Curation
Hongru Zhao, Jinwen Fu, Tuan Pham
https://arxiv.org/abs/2511.09002 https…

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 2724 nodes and 14322 edges.
Tags: Biological, Protein interactions, Unweighted…

ML-Asset Management: Curation, Discovery, and Utilization
Mengying Wang, Moming Duan, Yicong Huang, Chen Li, Bingsheng He, Yinghui Wu
https://arxiv.org/abs/2509.23577 https://…

ML-Asset Management: Curation, Discovery, and Utilization
Machine learning (ML) assets, such as models, datasets, and metadata, are central to modern ML workflows. Despite their explosive growth in practice, these assets are often underutilized due to fragmented documentation, siloed storage, inconsistent licensing, and lack of unified discovery mechanisms, making ML-asset management an urgent challenge. This tutorial offers a comprehensive overview of ML-asset management activities across its lifecycle, including curation, discovery, and utilization.…

We offer rolling admission for our online grad certificate in Digital Curation, so you can apply any time. Our spring courses will cover #DigitalPreservation #DigitalStorage #Metadata

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 912 nodes and 22738 edges.
Tags: Biological, Protein interactions, Unweighted<…

Back to the Future Museum -- Speculative Design for Virtual Citizen-Curated Museums
Richard Rhodes, Sandra Woolley
https://arxiv.org/abs/2510.06472 https://

Back to the Future Museum -- Speculative Design for Virtual Citizen-Curated Museums
This forward-looking paper uses speculative design fiction to explore future museum scenarios where citizen curators design and share immersive virtual reality museums populated with tangible heritage artefacts, intangible virtual elements and interactive experiences. The work also explores takeaway 'asset packs' containing 3D artefact models, curation assets, and interactive experiences, and we envisage a visit to the future museum, where the physical and virtual experiences interplay. Finally…

«The recent news concerning GigaScience’s owners, BGI, laying off the entire editorial, software and curation team in Hong Kong on short notice, has filled us with both surprise and deep disappointment.»
Indeed, GigaScience has always been a mainstay at BOSC and the larger intersection of life sciences & open source/science & citizen science.
https://www.open-bio.org/2025/09/30/2025-09-30-gigascience/

Mellum: Production-Grade in-IDE Contextual Code Completion with Multi-File Project Understanding
Nikita Pavlichenko, Iurii Nazarov, Ivan Dolgov, Ekaterina Garanina, Dmitry Ustalov, Ivan Bondyrev, Kseniia Lysaniuk, Evgeniia Vu, Kirill Chekmenev, Joseph Shtok, Yaroslav Golubev, Anton Semenkin, Uladzislau Sazanovich
https://arxiv.org/abs/2510…

Mellum: Production-Grade in-IDE Contextual Code Completion with Multi-File Project Understanding
We present the Mellum models family, open-weight code completion models designed for interactive use in JetBrains IDEs. Mellums have 4B parameters, adopt a Llama-style architecture, and are pre-trained on ~4T tokens of permissively licensed, multi-language code. Our studies show that (i) careful data curation and staged training significantly improve the model's quality, (ii) editor-critical capabilities such as context packing are necessary for high-quality suggestions, and (iii) a compact, ta…

LABELING COPILOT: A Deep Research Agent for Automated Data Curation in Computer Vision
Debargha Ganguly, Sumit Kumar, Ishwar Balappanawar, Weicong Chen, Shashank Kambhatla, Srinivasan Iyengar, Shivkumar Kalyanaraman, Ponnurangam Kumaraguru, Vipin Chaudhary
https://arxiv.org/abs/2509.22631

LABELING COPILOT: A Deep Research Agent for Automated Data Curation in Computer Vision
Curating high-quality, domain-specific datasets is a major bottleneck for deploying robust vision systems, requiring complex trade-offs between data quality, diversity, and cost when researching vast, unlabeled data lakes. We introduce Labeling Copilot, the first data curation deep research agent for computer vision. A central orchestrator agent, powered by a large multimodal language model, uses multi-step reasoning to execute specialized tools across three core capabilities: (1) Calibrated Di…

The 21st century is a quarter over! Isn't that terrifying? Even more terrifying if your library, archive, or museum is still stuck in the previous century.
That's why we started UMaine's online Digital Curation program, which teaches the digital skills necessary to bring your collection into the 21st. More at https://D…

OpenGVL - Benchmarking Visual Temporal Progress for Data Curation
Pawe{\l} Budzianowski, Emilia Wi\'snios, Gracjan G\'oral, Igor Kulakov, Viktor Petrenko, Krzysztof Walas
https://arxiv.org/abs/2509.17321

OpenGVL - Benchmarking Visual Temporal Progress for Data Curation
Data scarcity remains one of the most limiting factors in driving progress in robotics. However, the amount of available robotics data in the wild is growing exponentially, creating new opportunities for large-scale data utilization. Reliable temporal task completion prediction could help automatically annotate and curate this data at scale. The Generative Value Learning (GVL) approach was recently proposed, leveraging the knowledge embedded in vision-language models (VLMs) to predict task prog…

Here’s my 35th “Long Links” outing, curation of long-form offerings, which assume that nobody has time to read all this stuff but one or two of the pieces might brighten your day. This one is mostly political but some of the politics are from France and China. Plus a way-cool analytical history of blogging and a section labeled “wonderful things”.

> Companies that can find the courage to trust their people, who bet on caring and curation and small models as differentiators instead of just rebadging big company products will have better choices and they’ll make better choices.
Among the bangers here from @… :
Pattern Recognition And Repetition | blarg

@… https://fluxblog.substack.com/

Fluxblog: Music Blog | Matthew Perpetua | Substack
The longest-running music blog in the game! This newsletter is focused on music curation - reviews of new music, playlists for Spotify/Apple/YouTube covering a wide range of genres and moments in history, and links to other cool things. Click to read Fluxblog: Music Blog, by Matthew Perpetua, a Substack publication with thousands of subscribers.

https://www.theargumentmag.com/p/treat-big-tech-like-big-tobacco
I expected to hate this take but actually I think it's great. Curation should be speech, not hosting. Plan accordingly.

Deconstructing Self-Bias in LLM-generated Translation Benchmarks
Wenda Xu, Sweta Agrawal, Vil\'em Zouhar, Markus Freitag, Daniel Deutsch
https://arxiv.org/abs/2509.26600 htt…

Deconstructing Self-Bias in LLM-generated Translation Benchmarks
As large language models (LLMs) begin to saturate existing benchmarks, automated benchmark creation using LLMs (LLM as a benchmark) has emerged as a scalable alternative to slow and costly human curation. While these generated test sets have to potential to cheaply rank models, we demonstrate a critical flaw. LLM generated benchmarks systematically favor the model that created the benchmark, they exhibit self bias on low resource languages to English translation tasks. We show three key finding…

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 759 nodes and 1593 edges.
Tags: Biological, Protein interactions, Unweighted

November is digital heritage month; in case you need help with your collection, UMaine's all-online graduate certificate has been called "a national standard for the study of digital curation." You can register now for my spring course in preservation or apply for the full certificate at https://DigitalCuration.UMaine.edu<…

Bayesian Anomaly Detection for Ia Cosmology: Automating SALT3 Data Curation
S. A. K. Leeney, W. J. Handley, H. T. J. Bevins, E. de Lera Acedo
https://arxiv.org/abs/2509.13394 ht…

Bayesian Anomaly Detection for Ia Cosmology: Automating SALT3 Data Curation
Precision cosmology with Type Ia supernovae (SNe Ia) requires robust quality control of large, heterogeneous datasets. Current data processing often relies on manual, subjective rejection of photometric data, a practice that is not scalable for forthcoming surveys like the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST). We present a Bayesian framework that automates this step by integrating anomaly detection directly into the light curve fitting process. While the framework …

Generative AI is slowly transforming digital curation, so we've integrated lessons on its technology, ethics, and techniques into our online graduate certificate at https://DigitalCuration.UMaine.edu.
Here's a related job as a Critical Artificial Intelligence Librarian at Dartmouth…

Cascade! Human in the loop shortcomings can increase the risk of failures in recommender systems
Wm. Matthew Kennedy, Nishanshi Shukla, Cigdem Patlak, Blake Chambers, Theodora Skeadas, Tuesday, Kingsley Owadara, Aayush Dhanotiya
https://arxiv.org/abs/2509.20099

Cascade! Human in the loop shortcomings can increase the risk of failures in recommender systems
Recommender systems are among the most commonly deployed systems today. Systems design approaches to AI-powered recommender systems have done well to urge recommender system developers to follow more intentional data collection, curation, and management procedures. So too has the "human-in-the-loop" paradigm been widely adopted, primarily to address the issue of accountability. However, in this paper, we take the position that human oversight in recommender system design also entails novel risk…

Crosslisted article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[12/21]:
- Efficient Multi-turn RL for GUI Agents via Decoupled Training and Adaptive Data Curation
Li, Hu, Shang, Wu, Liu, Liu, Gao, Shi, Zhang, Zhang, Shi, YU, Wu, Wu, Jia, Xiang, He, Li

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 2436 nodes and 136930 edges.
Tags: Biological, Protein interactions, Unweighte…

Identifying Constructive Conflict in Online Discussions through Controversial yet Toxicity Resilient Posts
Ozgur Can Seckin, Bao Tran Truong, Alessandro Flammini, Filippo Menczer
https://arxiv.org/abs/2509.18303

Experiversum: an Ecosystem for Curating and Enhancing Data-Driven Experimental Science
Genoveva Vargas-Solar (LIRIS), Umberto Costa (ERIC, UL2), J\'er\^ome Darmont (ERIC, UL2), Javier Espinosa-Oviedo (ERIC, UCBL), Carmem Hara (ERIC, UL2), Sabine Loudcher (ERIC, UL2), Regina Motz, Martin A. Musicante, Jos\'e-Luis Zechinelli-Martini
https://

Experiversum: an Ecosystem for Curating and Enhancing Data-Driven Experimental Science
This paper introduces Experiversum, a lakehouse-based ecosystem that supports the curation, documentation and reproducibility of exploratory experiments. Experiversum enables structured research through iterative data cycles, while capturing metadata and collaborative decisions. Demonstrated through case studies in Earth, Life and Political Sciences, Experiversum promotes transparent workflows and multi-perspective result interpretation. Experiversum bridges exploratory and reproducible researc…

Crosslisted article(s) found for astro-ph.CO. https://arxiv.org/list/astro-ph.CO/new
[1/1]:
- Bayesian Anomaly Detection for Ia Cosmology: Automating SALT3 Data Curation
S. A. K. Leeney, W. J. Handley, H. T. J. Bevins, E. de Lera Acedo

Learning Relationships Between Separate Audio Tracks for Creative Applications
Balthazar Bujard (IRCAM, SU), J\'er\^ome Nika (IRCAM), F\'ed\'eric Bevilacqua (IRCAM), Nicolas Obin
https://arxiv.org/abs/2509.25296

Learning Relationships Between Separate Audio Tracks for Creative Applications
This paper presents the first step in a research project situated within the field of musical agents. The objective is to achieve, through training, the tuning of the desired musical relationship between a live musical input and a real-time generated musical output, through the curation of a database of separated tracks. We propose an architecture integrating a symbolic decision module capable of learning and exploiting musical relationships from such musical corpus. We detail an offline implem…

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 2436 nodes and 136930 edges.
Tags: Biological, Protein interactions, Unweighte…

Building Data-Driven Occupation Taxonomies: A Bottom-Up Multi-Stage Approach via Semantic Clustering and Multi-Agent Collaboration
Nan Li, Bo Kang, Tijl De Bie
https://arxiv.org/abs/2509.15786

Building Data-Driven Occupation Taxonomies: A Bottom-Up Multi-Stage Approach via Semantic Clustering and Multi-Agent Collaboration
Creating robust occupation taxonomies, vital for applications ranging from job recommendation to labor market intelligence, is challenging. Manual curation is slow, while existing automated methods are either not adaptive to dynamic regional markets (top-down) or struggle to build coherent hierarchies from noisy data (bottom-up). We introduce CLIMB (CLusterIng-based Multi-agent taxonomy Builder), a framework that fully automates the creation of high-quality, data-driven taxonomies from raw job …

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 1237 nodes and 1735 edges.
Tags: Biological, Protein interactions, Unweighted<…

SAIL-VL2 Technical Report
Weijie Yin, Yongjie Ye, Fangxun Shu, Yue Liao, Zijian Kang, Hongyuan Dong, Haiyang Yu, Dingkang Yang, Jiacong Wang, Han Wang, Wenzhuo Liu, Xiao Liang, Shuicheng Yan, Chao Feng
https://arxiv.org/abs/2509.14033

SAIL-VL2 Technical Report
We introduce SAIL-VL2, an open-suite vision-language foundation model (LVM) for comprehensive multimodal understanding and reasoning. As the successor to SAIL-VL, SAIL-VL2 achieves state-of-the-art performance at the 2B and 8B parameter scales across diverse image and video benchmarks, demonstrating strong capabilities from fine-grained perception to complex reasoning. Three core innovations drive its effectiveness. First, a large-scale data curation pipeline with scoring and filtering strategi…

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 431 nodes and 438 edges.
Tags: Biological, Protein interactions, Unweighted

celegans_interactomes: C. elegans interactomes (2009)
Ten networks of protein-protein interactions in Caenorhabditis elegans (nematode), from yeast two-hybrid experiments, biological process maps, literature curation, orthologous interactions, and genetic interactions. The WI8 network combines WI2004, WI2007 and BPmaps, while the Integrated Network combines data from all sources.
This network has 2724 nodes and 14322 edges.
Tags: Biological, Protein interactions, Unweighted…