Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
mastodon.gamedev.place is one of the many independent Mastodon servers you can use to participate in the fediverse.
Mastodon server focused on game development and related topics.

Administered by:

Server stats:

5.1K
active users

mastodon.gamedev.place: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.8

#compsci

40 posts20 participants1 post today
Public
Emma Stamm

something I somehow hadn't thought about directly until just now- the influence of #arXiv , a non-peer-reviewed platform, on perception of #AI (and #compsci in general). i know arXiv has some refereeing system but i don't know how rigorous it is.

Public
Compsci Weekly

Absolute Zero: Reinforced Self-play Reasoning with Zero Data [R]

arxiv.org/abs/2505.03335

Discussions: discu.eu/q/https://www.arxiv.o

arXiv logo
arXiv.orgAbsolute Zero: Reinforced Self-play Reasoning with Zero DataReinforcement learning with verifiable rewards (RLVR) has shown promise in enhancing the reasoning capabilities of large language models by learning directly from outcome-based rewards. Recent RLVR works that operate under the zero setting avoid supervision in labeling the reasoning process, but still depend on manually curated collections of questions and answers for training. The scarcity of high-quality, human-produced examples raises concerns about the long-term scalability of relying on human supervision, a challenge already evident in the domain of language model pretraining. Furthermore, in a hypothetical future where AI surpasses human intelligence, tasks provided by humans may offer limited learning potential for a superintelligent system. To address these concerns, we propose a new RLVR paradigm called Absolute Zero, in which a single model learns to propose tasks that maximize its own learning progress and improves reasoning by solving them, without relying on any external data. Under this paradigm, we introduce the Absolute Zero Reasoner (AZR), a system that self-evolves its training curriculum and reasoning ability by using a code executor to both validate proposed code reasoning tasks and verify answers, serving as an unified source of verifiable reward to guide open-ended yet grounded learning. Despite being trained entirely without external data, AZR achieves overall SOTA performance on coding and mathematical reasoning tasks, outperforming existing zero-setting models that rely on tens of thousands of in-domain human-curated examples. Furthermore, we demonstrate that AZR can be effectively applied across different model scales and is compatible with various model classes.
#compsci#machinelearning
Public
Compsci Weekly

IBM AI Releases Granite 4.0 Tiny Preview: A Compact Open-Language Model Optimized for Long-Context and Instruction Tasks

marktechpost.com/2025/05/03/ib

Discussions: discu.eu/q/https://www.marktec

MarkTechPost · IBM AI Releases Granite 4.0 Tiny Preview: A Compact Open-Language Model Optimized for Long-Context and Instruction TasksIBM AI Releases Granite 4.0 Tiny Preview: A Compact Open-Language Model Optimized for Long-Context and Instruction Tasks
#compsci#machinelearning
Public
mempko

Rust does seem to have a lot of nice features. My biggest blocker for me going to Rust from C++ is that C++ has much better support for generic programming. And now that Concepts have landed, I'm not aware of any language that can compete in this area.

#rust#cpp#CPlusCplus
Public
Compsci Weekly

GenAI-accelerated TLA+ challenge

foundation.tlapl.us/challenge/

Discussions: discu.eu/q/https://foundation.

TLA+ FoundationGenAI-accelerated TLA+ challenge The TLA+ Foundation, in collaboration with NVIDIA, is pleased to announce the GenAI-accelerated TLA+ challenge—an open call for submissions that explore the intersection of TLA+ and generative AI. This initiative aims to foster practical and innovative tooling, workflows, and approaches that bring the capabilities of generative AI and LLMs to TLA+. Participants are invited to develop engineering-oriented solutions that advance the usability, accessibility, and automation of formal specification through the integration of GenAI. Awards 1st Place: Nvidia GeForce RTX 5090 (sponsored by NVIDIA) 2nd Place: One-year single seat, individual subscription to Github Copilot Pro+ (sponsored by the TLA+ Foundation) 3rd Place: One-year single seat, individual subscription to Github Copilot Pro (sponsored by the TLA+ Foundation) Example Project Areas Participants may submit work including, but not limited to:
#compsci#formalmethods#vibecoding
Public
Compsci Weekly

Stop treating `AGI' as the north-star goal of AI research

arxiv.org/abs/2502.03689

Discussions: discu.eu/q/https://arxiv.org/a

arXiv logo
arXiv.orgStop treating `AGI' as the north-star goal of AI researchThe AI research community plays a vital role in shaping the scientific, engineering, and societal goals of AI research. In this position paper, we argue that focusing on the highly contested topic of `artificial general intelligence' (`AGI') undermines our ability to choose effective goals. We identify six key traps -- obstacles to productive goal setting -- that are aggravated by AGI discourse: Illusion of Consensus, Supercharging Bad Science, Presuming Value-Neutrality, Goal Lottery, Generality Debt, and Normalized Exclusion. To avoid these traps, we argue that the AI research community needs to (1) prioritize specificity in engineering and societal goals, (2) center pluralism about multiple worthwhile approaches to multiple valuable goals, and (3) foster innovation through greater inclusion of disciplines and communities. Therefore, the AI research community needs to stop treating `AGI' as the north-star goal of AI research.
#compsci#machinelearning
ExploreLive feeds
About