With the proliferation of data across various domains, there is a critical demand for tools that enable non-experts to derive meaningful insights without deep data analysis skills. To address this need, existing automatic fact sheet generation tools offer heuristic-based solutions to extract facts and generate stories. However, they inadequately grasp the semantics of data and struggle to generate narratives that fully capture the semantics of the dataset or align the fact sheet with specific user needs. Addressing these shortcomings, this paper introduces \tool, a novel tool designed for the automatic generation and customisation of fact sheets. \tool applies the concept of collaborative AI workers to transform raw tabular dataset into comprehensive, visually compelling fact sheets. We define effective taxonomy to profile AI worker for specialised tasks. Furthermore, \tool empowers users to refine these fact sheets through intuitive natural language commands, ensuring the final outputs align closely with individual preferences and requirements. Our user evaluation with 18 participants confirms that \tool not only surpasses state-of-the-art baselines in automated fact sheet production but also provides a positive user experience during customization tasks.

How can we verify whether copyrighted content was used to train a large vision-language model (VLM) without direct access to its training data? Motivated by the hypothesis that a VLM is able to recognize images from its training corpus, we propose DIS-CO, a novel approach to infer the inclusion of copyrighted content during the model's development. By repeatedly querying a VLM with specific frames from targeted copyrighted material, DIS-CO extracts the content's identity through free-form text completions. To assess its effectiveness, we introduce MovieTection, a benchmark comprising 14,000 frames paired with detailed captions, drawn from films released both before and after a model's training cutoff. Our results show that DIS-CO significantly improves detection performance, nearly doubling the average AUC of the best prior method on models with logits available. Our findings also highlight a broader concern: all tested models appear to have been exposed to some extent to copyrighted content. Our code and data are available at https://github.com/avduarte333/DIS-CO

A couple of years ago, Grimes thought art might be dying. She worried that TikTok was overwhelming attention spans; that transgressive artists were becoming more sanitized; that gimmicky NFTs like the Bored Ape Yacht Club--digital cartoon monkeys which were selling for millions of dollars--were warping value systems. "I just went through this whole big'art isn't worth anything' internal existential crisis," the Canadian singer-songwriter says. "But I've come out the other end thinking, actually, maybe it's the main thing that matters. In the last year, I feel like things became way more about artists again." The rise of AI, Grimes believes, has played a role in that shift, perhaps paradoxically. Earlier this month, Grimes was honored at the TIME100 AI Impact Awards in Dubai for her role in shaping the present and future of the technology. While many other artists are terrified of AI and its potential to replace them, Grimes has embraced the technology, even releasing an AI tool allowing people to sing through her voice. Grimes' penchant for seriously engaging with what others fear or distrust makes her one of pop culture's most singular--and at times divisive--figures. But Grimes wears her contrarianism as a badge of honor, and doesn't hesitate to offer insights and perspectives on a variety of issues. "I'm so canceled that I basically have nothing left to lose," she says. She argues that hyper-partisan hysteria has consumed social media, and wishes people would have more measured, nuanced conversations, even with people that they disagree with. "A lot of people think I'm one way or the other, but my whole vibe is just like, I just want people to think well," she says.

From making tea to cleaning the floors, everyday life often feels like one huge chore. But the opportunity to offload such menial tasks to your own personal robot helper may arrive sooner than you think. In a promo clip, the advanced humanoid boils the kettle, vacuums floors, carries groceries, cleans windows and puts up a picture frame. At the end of the video, it takes a well-earned sit in the longue – while its blissfully-happy owners drink wine in the next room. Although it is currently a prototype, the creation could be autonomously completing chores in customers' homes by the end of the decade.

When Xi Jinping, China's leader, made his entrance at a symposium with a group of top entrepreneurs this past week, he seemed to be in good spirits. China has had a few good weeks. Artificial intelligence models by the startup DeepSeek sent U.S. stocks tumbling and Western commentators screaming, "Sputnik moment." Then, an animated film based on Chinese mythology raked in nearly 2 billion. Xi signaled that he stood behind the private sector at the meeting on Monday, pushing the Hong Kong stock market to its highest point in three years.

There have recently been many cases of unverified or misleading information circulating quickly over bogus web networks and news portals. This false news creates big damage to society and misleads people. For Example, in 2019, there was a rumor that the Padma Bridge of Bangladesh needed 100,000 human heads for sacrifice. This rumor turns into a deadly position and this misleading information takes the lives of innocent people. There is a lot of work in English but a few works in Bangla. In this study, we are going to identify the fake news from the unconsidered news source to provide the newsreader with natural news or real news. The paper is based on the combination of convolutional neural network (CNN) and long short-term memory (LSTM), where CNN is used for deep feature extraction and LSTM is used for detection using the extracted feature. The first thing we did to deploy this piece of work was data collection. We compiled a data set from websites and attempted to deploy it using the methodology of deep learning which contains about 50k of news. With the proposed model of Multichannel combined CNN-LSTM architecture, our model gained an accuracy of 75.05%, which is a good sign for detecting fake news in Bangla.

This paper argues that generating output tokens is more effective than using pooled representations for prediction tasks because token-level generation retains more mutual information. Since LLMs are trained on massive text corpora using next-token prediction, generation aligns naturally with their learned behavior. Using the Data Processing Inequality (DPI), we provide both theoretical and empirical evidence supporting this claim. However, autoregressive models face two key challenges when used for prediction: (1) exposure bias, where the model sees ground truth tokens during training but relies on its own predictions during inference, leading to errors, and (2) format mismatch, where discrete tokens do not always align with the tasks required output structure. To address these challenges, we introduce PredGen(Predicting Through Generating), an end to end framework that (i) uses scheduled sampling to reduce exposure bias, and (ii) introduces a task adapter to convert the generated tokens into structured outputs. Additionally, we introduce Writer-Director Alignment Loss (WDAL), which ensures consistency between token generation and final task predictions, improving both text coherence and numerical accuracy. We evaluate PredGen on multiple classification and regression benchmarks. Our results show that PredGen consistently outperforms standard baselines, demonstrating its effectiveness in structured prediction tasks.

Tip-of-the-tongue (TOT) search occurs when a user struggles to recall a specific identifier, such as a document title. While common, existing search systems often fail to effectively support TOT scenarios. Research on TOT retrieval is further constrained by the challenge of collecting queries, as current approaches rely heavily on community question-answering (CQA) websites, leading to labor-intensive evaluation and domain bias. To overcome these limitations, we introduce two methods for eliciting TOT queries - leveraging large language models (LLMs) and human participants - to facilitate simulated evaluations of TOT retrieval systems. Our LLM-based TOT user simulator generates synthetic TOT queries at scale, achieving high correlations with how CQA-based TOT queries rank TOT retrieval systems when tested in the Movie domain. Additionally, these synthetic queries exhibit high linguistic similarity to CQA-derived queries. For human-elicited queries, we developed an interface that uses visual stimuli to place participants in a TOT state, enabling the collection of natural queries. In the Movie domain, system rank correlation and linguistic similarity analyses confirm that human-elicited queries are both effective and closely resemble CQA-based queries. These approaches reduce reliance on CQA-based data collection while expanding coverage to underrepresented domains, such as Landmark and Person. LLM-elicited queries for the Movie, Landmark, and Person domains have been released as test queries in the TREC 2024 TOT track, with human-elicited queries scheduled for inclusion in the TREC 2025 TOT track. Additionally, we provide source code for synthetic query generation and the human query collection interface, along with curated visual stimuli used for eliciting TOT queries.

The Musical Instrument Digital Interface (MIDI), introduced in 1983, revolutionized music production by allowing computers and instruments to communicate efficiently. MIDI files encode musical instructions compactly, facilitating convenient music sharing. They benefit Music Information Retrieval (MIR), aiding in research on music understanding, computational musicology, and generative music. The GigaMIDI dataset contains over 1.4 million unique MIDI files, encompassing 1.8 billion MIDI note events and over 5.3 million MIDI tracks. GigaMIDI is currently the largest collection of symbolic music in MIDI format available for research purposes under fair dealing. Distinguishing between non-expressive and expressive MIDI tracks is challenging, as MIDI files do not inherently make this distinction. To address this issue, we introduce a set of innovative heuristics for detecting expressive music performance. These include the Distinctive Note Velocity Ratio (DNVR) heuristic, which analyzes MIDI note velocity; the Distinctive Note Onset Deviation Ratio (DNODR) heuristic, which examines deviations in note onset times; and the Note Onset Median Metric Level (NOMML) heuristic, which evaluates onset positions relative to metric levels. Our evaluation demonstrates these heuristics effectively differentiate between non-expressive and expressive MIDI tracks. Furthermore, after evaluation, we create the most substantial expressive MIDI dataset, employing our heuristic, NOMML. This curated iteration of GigaMIDI encompasses expressively-performed instrument tracks detected by NOMML, containing all General MIDI instruments, constituting 31% of the GigaMIDI dataset, totalling 1,655,649 tracks.

Gestures are an integral part of non-verbal communication, with meanings that vary across cultures, and misinterpretations that can have serious social and diplomatic consequences. As AI systems become more integrated into global applications, ensuring they do not inadvertently perpetuate cultural offenses is critical. To this end, we introduce Multi-Cultural Set of Inappropriate Gestures and Nonverbal Signs (MC-SIGNS), a dataset of 288 gesture-country pairs annotated for offensiveness, cultural significance, and contextual factors across 25 gestures and 85 countries. Through systematic evaluation using MC-SIGNS, we uncover critical limitations: text-to-image (T2I) systems exhibit strong US-centric biases, performing better at detecting offensive gestures in US contexts than in non-US ones; large language models (LLMs) tend to over-flag gestures as offensive; and vision-language models (VLMs) default to US-based interpretations when responding to universal concepts like wishing someone luck, frequently suggesting culturally inappropriate gestures. These findings highlight the urgent need for culturally-aware AI safety mechanisms to ensure equitable global deployment of AI technologies.