Media
TEXT2DB: Integration-Aware Information Extraction with Large Language Model Agents
Jiao, Yizhu, Li, Sha, Zhou, Sizhe, Ji, Heng, Han, Jiawei
The task of information extraction (IE) is to extract structured knowledge from text. However, it is often not straightforward to utilize IE output due to the mismatch between the IE ontology and the downstream application needs. We propose a new formulation of IE TEXT2DB that emphasizes the integration of IE output and the target database (or knowledge base). Given a user instruction, a document set, and a database, our task requires the model to update the database with values from the document set to satisfy the user instruction. This task requires understanding user instructions for what to extract and adapting to the given DB/KB schema for how to extract on the fly. To evaluate this new task, we introduce a new benchmark featuring common demands such as data infilling, row population, and column addition. In addition, we propose an LLM agent framework OPAL (Observe-PlanAnalyze LLM) which includes an Observer component that interacts with the database, the Planner component that generates a code-based plan with calls to IE models, and the Analyzer component that provides feedback regarding code quality before execution. Experiments show that OPAL can successfully adapt to diverse database schemas by generating different code plans and calling the required IE models. We also highlight difficult cases such as dealing with large databases with complex dependencies and extraction hallucination, which we believe deserve further investigation. Source code: https://github.com/yzjiao/Text2DB
Learning Interpretable Features in Audio Latent Spaces via Sparse Autoencoders
Paek, Nathan, Zang, Yongyi, Yang, Qihui, Leistikow, Randal
While sparse autoencoders (SAEs) successfully extract interpretable features from language models, applying them to audio generation faces unique challenges: audio's dense nature requires compression that obscures semantic meaning, and automatic feature characterization remains limited. We propose a framework for interpreting audio generative models by mapping their latent representations to human-interpretable acoustic concepts. We train SAEs on audio autoencoder latents, then learn linear mappings from SAE features to discretized acoustic properties (pitch, amplitude, and timbre). This enables both controllable manipulation and analysis of the AI music generation process, revealing how acoustic properties emerge during synthesis. We validate our approach on continuous (DiffRhythm-VAE) and discrete (EnCodec, WavTokenizer) audio latent spaces, and analyze DiffRhythm, a state-of-the-art text-to-music model, to demonstrate how pitch, timbre, and loudness evolve throughout generation. While our work is only done on audio modality, our framework can be extended to interpretable analysis of visual latent space generation models.
Verdicts in as Liam Hemsworth takes over as The Witcher
The latest season of Netflix's The Witcher has landed - with one big difference. Former lead actor Henry Cavill has been replaced as main character Geralt of Rivia by Liam Hemsworth. The Australian has stepped in for the final two seasons of the fantasy show, based on a popular series of novels and video games. Previously, British actor Cavill had portrayed the title character, a monster hunter with supernatural abilities known as the White Wolf. When he announced he was passing the torch to Hemsworth in October 2022, describing him as a fantastic actor, not all fans agreed.
Listen up: The Popular Science 'Ask Us Anything' podcast is back
Science Announcements Listen up: The Popular Science'Ask Us Anything' podcast is back Breakthroughs, discoveries, and DIY tips sent every weekday. Why do we have toenails? How do airplane toilets actually work? For more than 150 years, has answered your questions--from the serious to the outlandish. Based on our wildly popular written series of the same name, the audio version features host Sarah Durn and the editors discussing everything from goose bumps to human composting.
If You Hated 'A House of Dynamite,' Watch This Classic Nuclear Thriller Instead
At a time when nuclear threats feel more alarming than ever, Netflix's doomsday film falls frustratingly flat. A 1964 masterpiece tells a much better cautionary tale. Somewhere over the Arctic reaches of North America, a nuclear bomber flies in a squadron, awaiting its orders. When a secret code appears on a machine in the cockpit, the crew looks at each other, stunned. The code is instructing them to attack.
The Download: Introducing: the new conspiracy age
Everything is a conspiracy theory now. Conspiracists are all over the White House, turning fringe ideas into dangerous policy. America's institutions are crumbling under the weight of deep suspicion and the lasting effects of covid isolation. Online echo chambers are getting harder to escape, and generative AI is altering the fabric of truth. A mix of technology and politics has given an unprecedented boost to once-fringe ideas--but they are pretty much the same fantasies that have been spreading for hundreds of years. MIT Technology Review helps break down how this moment is changing science and technology--and how we can make it through.
Four thoughts from Bill Gates on climate tech
Why he thinks near-term targets can be a distraction, and what technologies he expects to power our future grid. Bill Gates doesn't shy away or pretend modesty when it comes to his stature in the climate world today. "Well, who's the biggest funder of climate innovation companies?" he asked a handful of journalists at a media roundtable event last week. "If there's someone else, I've never met them." The former Microsoft CEO has spent the last decade investing in climate technology through Breakthrough Energy, which he founded in 2015. Ahead of the UN climate meetings kicking off next week, Gates published a memo outlining what he thinks activists and negotiators should focus on and how he's thinking about the state of climate tech right now.