I'm almost embarrassed to admit that the first time I saw one of Meta's new holographic avatars today, during a WhatsApp call with a Meta rep I had just met a few minutes before, I didn't immediately clock that I was speaking to an avatar. That's how far the company's avatar technology has come in the last couple years. Meta announced at Connect that it's bringing its holographic avatars to its new VR glasses and to the Meta Ray-Ban Display frames. I got the chance to experience both ahead of the keynote and was surprised at how far the tech has come since the first time I experienced it on Meta's Orion prototype. Setting up an avatar is a bit different depending on which device you're using.

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.

Plus, a timely soccer-related management sim that doesn't hit the back of the net. Welcome to our latest roundup of what's going on in the indie game space. As ever, we've got some new releases and news of upcoming games to share with you. However, it's Steam Next Fest time, so there are literally thousands of fresh game demos to jump into. One of the main aims of Steam Next Fest is to help players discover interesting games.

Accurate motion forecasting for traffic agents is crucial for ensuring the safety and efficiency of autonomous driving systems in dynamically changing environments. Mainstream methods adopt a one-query-one-trajectory paradigm, where each query corresponds to a unique trajectory for predicting multi-modal trajectories. While straightforward and effective, the absence of detailed representation of future trajectories may yield suboptimal outcomes, given that the agent states dynamically evolve over time. To address this problem, we introduce DeMo, a framework that decouples multi-modal trajectory queries into two types: mode queries capturing distinct directional intentions and state queries tracking the agent's dynamic states over time.

Nvidia's new AI upscaling gaming technology struck gamers as uncanny and off-putting. Developers don't seem to like it, either, but it could be "the default" in a few years. Nvidia announced a new version of its DLSS AI upscaling technology for its graphics cards earlier this week at its GPU Technology Conference (GTC), which it calls the Super Bowl of AI . But unlike previous versions of DLSS that used AI to improve frame rates in video games, DLSS 5 has a much more ambitious calling: using generative AI to make character faces in games look more realistic and detailed. The demonstration received sharp blowback on social media, with many finding the effect off-putting, reacting with outright disgust, and calling it yet another example of AI slop .

Li Zhang (lizhangfd@fudan.edu.cn) is the corresponding author. Previous methods, as depicted in (a), use only one mode query for each trajectory.

Samsung Galaxy Unpacked 2026 is Feb. 25 Valve's Steam Machine: Everything we know Plus, a tactical shooter from the team behind Half-Life remake Black Mesa. Welcome to our latest roundup of what's going on in the indie game space. A whole bunch of compelling games arrived this week, and Sony dropped some news about more that are on the way to PS5 and other platforms during its State of Play stream on Thursday . For one thing, I didn't have a prequel for, one of my favorite games of the last few years on my bingo card. It's really neat that Motion Twin and Evil Empire -- the studios behind and its expansions, respectively -- are getting to make a proper Castlevania game .

Detecting Alzheimer's disease (AD) from narrative transcripts challenges large language models (LLMs): pre-training rarely covers this out-of-distribution task, and all transcript demos describe the same scene, producing highly homogeneous contexts. These factors cripple both the model's built-in task knowledge (\textbf{task cognition}) and its ability to surface subtle, class-discriminative cues (\textbf{contextual perception}). Because cognition is fixed after pre-training, improving in-context learning (ICL) for AD detection hinges on enriching perception through better demonstration (demo) sets. We demonstrate that standard ICL quickly saturates, its demos lack diversity (context width) and fail to convey fine-grained signals (context depth), and that recent task vector (TV) approaches improve broad task adaptation by injecting TV into the LLMs' hidden states (HSs), they are ill-suited for AD detection due to the mismatch of injection granularity, strength and position. To address these bottlenecks, we introduce \textbf{DA4ICL}, a demo-centric anchoring framework that jointly expands context width via \emph{\textbf{Diverse and Contrastive Retrieval}} (DCR) and deepens each demo's signal via \emph{\textbf{Projected Vector Anchoring}} (PVA) at every Transformer layer. Across three AD benchmarks, DA4ICL achieves large, stable gains over both ICL and TV baselines, charting a new paradigm for fine-grained, OOD and low-resource LLM adaptation.