Deep learning, particularly with the advancement of Large Language Models, has transformed biomolecular modeling, with protein advances (e.g., ESM) inspiring emerging RNA language models such as RiNALMo. Yet how and what these RNA Language Models internally encode about messenger RNA (mRNA) or non-coding RNA (ncRNA) families remains unclear. We present SAE- RNA, interpretability model that analyzes RiNALMo representations and maps them to known human-level biological features. Our work frames RNA interpretability as concept discovery in pretrained embeddings, without end-to-end retraining, and provides practical tools to probe what RNA LMs may encode about ncRNA families. The model can be extended to close comparisons between RNA groups, and supporting hypothesis generation about previously unrecognized relationships.

Concurrent multi-player mean-payoff games are important models for systems of agents with individual, non-dichotomous preferences. Whilst these games have been extensively studied in terms of their equilibria in non-cooperative settings, this paper explores an alternative solution concept: the core from cooperative game theory. This concept is particularly relevant for cooperative AI systems, as it enables the modelling of cooperation among agents, even when their goals are not fully aligned. Our contribution is twofold. First, we provide a characterisation of the core using discrete geometry techniques and establish a necessary and sufficient condition for its non-emptiness. We then use the characterisation to prove the existence of polynomial witnesses in the core. Second, we use the existence of such witnesses to solve key decision problems in rational verification and provide tight complexity bounds for the problem of checking whether some/every equilibrium in a game satisfies a given LTL or GR(1) specification. Our approach is general and can be adapted to handle other specifications expressed in various fragments of LTL without incurring additional computational costs.

The COVID-19 pandemic has fueled efforts to integrate artificial intelligence into the healthcare ecosystem. Those efforts have so far achieved mixed results, according to an investment survey that found AI startups with domain expertise outperforming enterprise AI leaders attempting to crack the AI healthcare market. The AI healthcare assessment released this week by KLAS Research found that "cross-industry AI giants" such as Amazon, Google, IBM and Microsoft are investing heavily in the healthcare sector. Of those, about half of survey respondents said Microsoft (NASDAQ: MSFT) offered the best AI tools, largely based on healthcare partnerships with domain experts. Indeed, those AI startups scored highest in terms of healthcare AI technology and client base.

Caption: Caption: In the Cybathlon, pilots with complete spinal cord injuries take part in a bike race with the help of functional electrical stimulation.ETH Zurich/Alessandro Della Bella Caption: Caption: In this team event, competitors use powered arm prosthesis to complete a series of tasks. Caption: Caption: Team Avalanche competes in the powered wheelchair race.ETH Zürich/Alessandro Della Bella Caption: Caption: Team Mahidol competes in the computer interface race.ETH Zürich/Nicola Pitaro Caption: Caption: Team Imperial GBR competes in the powered arm prosthesis race.ETH Zürich/Nicola Pitaro Caption: Caption: Team OssurPowerKnee competes in the powered leg prosthesis race.ETH Zürich/Nicola Pitaro Caption: Caption: A participant with limited mobility can climb steps with the help of an exoskeleton.ETH Zurich/Alessandro Della Bella Caption: Caption: Team Meltin competes in the functional electrical stimulation bike race.ETH Zurich/Nicola Pitaro Caption: Caption: Team Varileg competes in teh powered exoskeleton race.ETH Zürich/Alessandro Della Bella Caption: Caption: A competitor in the powered leg prosthesis race.ETH Zürich/Alessandro Della Bella In pop culture, cyborgs can fly, throw cars, and blow up buildings. Nobody did any of those things at the world's first-ever cyborg Olympics--the Cybathlon in Zurich, Switzerland, held earlier this month--but the action was just as miraculous for a different reason. Using the latest bionic technology, disabled competitors paired up with prosthetics developers to accomplish tasks ranging from bread slicing to bike racing. Of the 59 teams, these three triumphed and scored top marks.

The design of the best economic mechanism for Sponsored Search Auctions (SSAs) is a central task in computational mechanism design/game theory. Two open questions concern (i) the adoption of user models more accurate than the currently used one and (ii) the choice between Generalized Second Price auction (GSP) and Vickrey–Clark–Groves mechanism (VCG). In this paper, we provide some contributions to answer these questions. We study Price of Anarchy (PoA) and Price of Stability (PoS) over social welfare and auctioneer’s revenue of GSP w.r.t. the VCG when the users follow the famous cascade model. Furthermore, we provide exact, randomized, and approximate algorithms, showing that in real–world settings (Yahoo! Webscope A3 dataset, 10 available slots) optimal allocations can be found in less than 1s with up to 1,000 ads, and can be approximated in less than 20ms even with more than 1,000 ads with an average accuracy greater than 99%.