Goto

Collaborating Authors

 real world


Interview with Minji Park: humans in the loop to keep language models up to date

AIHub

Minji Park is a PhD candidate at the Korea Institute of Energy Technology (KENTECH). In her paper, with co-authors Seunghyun Yoon and Hyuk Lim, she introduces a framework which aims to help make large language models stay more up-to-date when pieces of information change. We hear from Minji about inspired her work, how the framework operates, and her plans to further develop this research. What is the topic of the research in your paper and why is it an interesting area for study? The paper is fun, because we are challenging the norm.


How Pures 24-hour chat timer forces daters to skip the endless banter

Mashable

Look Up Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Safety Net Gift Ideas For Everyone On Your List Mashable @Home In My Bag Versus Trending Now Say More Creator Playbook Mashable Selects Getting Technical with Mashable All Series How Pure's 24-hour chat timer forces daters to skip the endless banter How introducing a sense of urgency into its chat mechanic made Pure a better dating app. All products featured here are independently selected by our editors and writers. If you buy something through links on our site, Mashable may earn an affiliate commission. Deal pricing and availability subject to change after time of publication. Learn more about how we select deals .


From lab to life: Why translating AI advances to the real world is a major challenge

AIHub

Systems that perform well in controlled lab settings often fall short when deployed in homes, in hospitals or on city streets. Vision-based systems, for example, hold promise for improving safety, monitoring health and aiding rehabilitation. For example, a camera-based system could help a clinician assess a patient's gait or rehabilitation progress. In the future, an assistive robot in an older adult's home might need to recognize when the person is struggling to stand or reaching for an object. But a major challenge is ensuring these robots perform reliably beyond the lab, particularly in messy, unpredictable or low-visibility conditions. As a computer scientist who studies AI systems designed to understand human behavior - especially through visual cues like body movements and interactions with objects - I see three main challenges to taking AI from the lab to the real world: generalization, human behavior and resources.


These Researchers Made AI Drive a Toyota Corolla to Get In-N-Out

WIRED

Three engineers put GPT, Claude, and Grok in charge of a real car. Only one of them was successful. Aditya Ramabadran, Simon Mahns, and Tobias Gessler, three AI engineers at a startup called Axiom, recently fancied In-N-Out Burger for lunch. They weren't about to drive themselves, though. Sitting in a 2024 Toyota Corolla near the Bay Area restaurant's drive-thru lane, they opened a laptop and asked OpenAI's GPT-6 Astra to take the wheel.


Can tech really get you off your phone and into real life?

FOX News

This material may not be published, broadcast, rewritten, or redistributed. Quotes displayed in real-time or delayed by at least 15 minutes. Market data provided by Factset . Powered and implemented by FactSet Digital Solutions . Mutual Fund and ETF data provided by LSEG . Asked Siri to call your bank? Floating solar panels could save cities millions.


NVIDIA's DLSS 5 launches with NBA 2K27 on September 3, but there's little reason for excitement

Engadget

It's hard to recall a feature that's been as instantly reviled as NVIDIA's DLSS 5, a new version of its AI upscaling technology that can modify character models and environments to chase the specter of "photorealism." NVIDIA calls it progress, but the initial examples from Resident Evil: Requiem and Starfield were uncanny misfires. Now, NVIDIA is gearing up to finally launch DLSS 5 alongside NBA 2K27 on September 3. And judging from the wildly negative responses to recent leaks of the technology, it's going to face a tough crowd. I had serious doubts about DLSS 5 at first, but after seeing some remote demonstrations of the tech from Gamescom, my hackles have cooled a bit. Sure, it's an AI tool that could be used to obscenely yassify a fragile character suffering from PTSD, like it did for Resident Evil: Requiem's co-lead Grace.


What happens when AI runs out of pictures?

AIHub

What happens when AI runs out of pictures? A hospital may only ever collect a few dozen scans of a rare condition - for example, an unusual tumour. The radiology department wants software to flag this on a scan - not to replace the specialist, but so a hospital without one still gets their scan checked the same way. The clinicians know what they're looking for. Over a decade, the hospital might gather 40 confirmed cases.


AI is 'not smart' so what's next in artificial intelligence?

BBC News

AI is'not smart' so what's next in artificial intelligence? We don't have robots that are nearly as good at understanding the physical world as a rat, says Yann LeCun, one of the leading figures in the world of artificial intelligence. He worked at Facebook-owner, Meta, for a decade, where he was chief AI scientist, but left in 2025 and founded Advanced Machine Intelligence Labs (AMI Labs). His goal is to move AI beyond current systems like ChatGPT, Claude and Gemini. They have their uses, he says, but will never be able to tackle complicated situations in the real world, like getting a robot to do household chores.


Domain Adaptation for and Real Policy Co Training

Neural Information Processing Systems

Behavior cloning has shown promise for robot manipulation, but real-world demonstrations are costly to acquire at scale. While simulated data offers a scalable alternative, particularly with advances in automated demonstration generation, transferring policies to the real world is hampered by various simulation and real domain gaps. In this work, we propose a unified sim-and-real co-training framework for learning generalizable manipulation policies that primarily leverages simulation and only requires a few real-world demonstrations. Central to our approach is learning a domain-invariant, task-relevant feature space. Our key insight is that aligning the joint distributions of observations and their corresponding actions across domains provides a richer signal than aligning observations (marginals) alone. We achieve this by embedding an Optimal Transport (OT)-inspired loss within the co-training framework, and extend this to an Unbalanced OT framework to handle the imbalance between abundant simulation data and limited real-world examples. We validate our method on challenging manipulation tasks, showing it can leverage abundant simulation data to achieve up to a 30% improvement in the real-world success rate and even generalize to scenarios seen only in simulation.


What's in Common? Multimodal Models Hallucinate When Reasoning Across Scenes

Neural Information Processing Systems

Multimodal language models possess a remarkable ability to handle an open-vocabulary worth of objects. Yet the best models still suffer from hallucinations when reasoning about scenes in the real world, revealing a gap between their seemingly strong performance on existing perception benchmarks that are saturating and their reasoning in the real world. To address this gap, we build a novel benchmark of in-the-wild scenes that we call Common-O Bench with more than 10.5k examples using exclusively new images not found in web training data to avoid contamination, Common-O goes beyond just perception, inspired by cognitive tests for humans, to probe reasoning across scenes by asking ``what's in common?''. We evaluate leading multimodal language models, including models specifically trained to reason. We find that perceiving objects in single images is easy for most models, yet reasoning across scenes is very challenging even for the best models, including reasoning models. Despite saturating many leaderboards focusing on perception, the best performing model only achieves 35\% on Common-O Bench---and on Common-O Complex, consisting of more complex scenes, the best model achieves only 1\%. Curiously, we find models are more prone to hallucinate when similar objects are present in the scene, suggesting models may be relying on object co-occurrence seen during training. Among the models we evaluated, we found scale can provide modest improvements while models explicitly trained with multi-image inputs show bigger improvements, suggesting scaled multi-image training may offer promise. We make our benchmark publicly available to spur research into the challenge of hallucination when reasoning across scenes.