PlayGen-MoG: Framework for Diverse Multi-Agent Play Generation via Mixture-of-Gaussians Trajectory Prediction
arXiv:2604.02447v1 Announce Type: new Abstract: Multi-agent trajectory generation in team sports requires models that capture both the diversity of possible plays and realistic spatial coordination between players on plays. Standard generative approaches such as Conditional Variational Autoencoders (CVAE) and diffusion models struggle with this task, exhibiting posterior collapse or convergence to the dataset mean. Moreover, most trajectory prediction methods operate in a forecasting regime that requires multiple frames of observed history, limiting their use for play design where only the initial formation is available. We present PlayGen-MoG, an extensible framework for formation-conditioned play generation that addresses these challenges through three design choices: 1/ a Mixture-of-Gau
View PDF HTML (experimental)
Abstract:Multi-agent trajectory generation in team sports requires models that capture both the diversity of possible plays and realistic spatial coordination between players on plays. Standard generative approaches such as Conditional Variational Autoencoders (CVAE) and diffusion models struggle with this task, exhibiting posterior collapse or convergence to the dataset mean. Moreover, most trajectory prediction methods operate in a forecasting regime that requires multiple frames of observed history, limiting their use for play design where only the initial formation is available. We present PlayGen-MoG, an extensible framework for formation-conditioned play generation that addresses these challenges through three design choices: 1/ a Mixture-of-Gaussians (MoG) output head with shared mixture weights across all agents, where a single set of weights selects a play scenario that couples all players' trajectories, 2/ relative spatial attention that encodes pairwise player positions and distances as learned attention biases, and 3/ non-autoregressive prediction of absolute displacements from the initial formation, eliminating cumulative error drift and removing the dependence on observed trajectory history, enabling realistic play generation from a single static formation alone. On American football tracking data, PlayGen-MoG achieves 1.68 yard ADE and 3.98 yard FDE while maintaining full utilization of all 8 mixture components with entropy of 2.06 out of 2.08, and qualitatively confirming diverse generation without mode collapse.
Comments: 9 pages, 4 figures, 2 tables. Accepted to CVPRW 2026
Subjects:
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as: arXiv:2604.02447 [cs.CV]
(or arXiv:2604.02447v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2604.02447
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Kevin Song [view email] [v1] Thu, 2 Apr 2026 18:18:53 UTC (1,526 KB)
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
modelannounceavailable
RightNow AI Releases AutoKernel: An Open-Source Framework that Applies an Autonomous Agent Loop to GPU Kernel Optimization for Arbitrary PyTorch Models
Writing fast GPU code is one of the most grueling specializations in machine learning engineering. Researchers from RightNow AI want to automate it entirely. The RightNow AI research team has released AutoKernel, an open-source framework that applies an autonomous LLM agent loop to GPU kernel optimization for arbitrary PyTorch models. The approach is straightforward: give [ ] The post RightNow AI Releases AutoKernel: An Open-Source Framework that Applies an Autonomous Agent Loop to GPU Kernel Optimization for Arbitrary PyTorch Models appeared first on MarkTechPost .
Production RAG: From Anti-Patterns to Platform Engineering
RAG is a distributed system . It becomes clear when moving beyond demos into production. It consists of independent services such as ingestion, retrieval, inference, orchestration, and observability. Each component introduces its own latency, scaling characteristics, and failure modes, making coordination, observability, and fault tolerance essential. RAG flowchart In regulated environments such as banking, these systems must also satisfy strict governance, auditability, and change-control requirements aligned with standards like SOX and PCI DSS. This article builds on existing frameworks like 12 Factor Agents (Dex Horthy)¹ and Google’s 16 Factor App² by exploring key anti-patterns and introducing the pillars required to take a typical RAG pipeline to production. I’ve included code snippet
Word2Vec Explained: The Moment Words Became Relations
How models first learned meaning from context — and why that changed everything In the first post, we built the base layer: Text → Tokens → Numbers → (lots of math) → Tokens → Text In the second post, we stayed with the deeper question: Once words become numbers, how does meaning not disappear? We saw that the answer is not “because numbers are magical.” The answer is this: the numbers are learned in a space that preserves relationships. That was the real story of embeddings. Now we are ready for the next step. Because once you accept that words can become numbers without losing meaning, the next question becomes unavoidable: How are those numbers actually learned? This is where Word2Vec enters the story. And Word2Vec matters for more than historical reasons. It was not just a clever neura
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Releases
Simple parallel estimation of the partition ratio for Gibbs distributions
arXiv:2505.18324v2 Announce Type: replace-cross Abstract: We consider the problem of estimating the partition function $Z(\beta)=\sum_x \exp(\beta(H(x))$ of a Gibbs distribution with the Hamiltonian $H:\Omega\rightarrow\{0\}\cup[1,n]$. As shown in [Harris & Kolmogorov 2024], the log-ratio $q=\ln (Z(\beta_{\max})/Z(\beta_{\min}))$ can be estimated with accuracy $\epsilon$ using $O(\frac{q \log n}{\epsilon^2})$ calls to an oracle that produces a sample from the Gibbs distribution for parameter $\beta\in[\beta_{\min},\beta_{\max}]$. That algorithm is inherently sequential, or {\em adaptive}: the queried values of $\beta$ depend on previous samples. Recently, [Liu, Yin & Zhang 2024] developed a non-adaptive version that needs $O( q (\log^2 n) (\log q + \log \log n + \epsilon^{-2}) )$ samples.
Polynomial-Time Almost Log-Space Tree Evaluation by Catalytic Pebbling
arXiv:2604.02606v1 Announce Type: cross Abstract: The Tree Evaluation Problem ($\mathsf{TreeEval}$) is a computational problem originally proposed as a candidate to prove a separation between complexity classes $\mathsf{P}$ and $\mathsf{L}$. Recently, this problem has gained significant attention after Cook and Mertz (STOC 2024) showed that $\mathsf{TreeEval}$ can be solved using $O(\log n\log\log n)$ bits of space. Their algorithm, despite getting very close to showing $\mathsf{TreeEval} \in \mathsf{L}$, falls short, and in particular, it does not run in polynomial time. In this work, we present the first polynomial-time, almost logarithmic-space algorithm for $\mathsf{TreeEval}$. For any $\varepsilon>0$, our algorithm solves $\mathsf{TreeEval}$ in time $\mathrm{poly}(n)$ while using $O(\
Near-Optimal Space Lower Bounds for Streaming CSPs
arXiv:2604.01400v1 Announce Type: cross Abstract: In a streaming constraint satisfaction problem (streaming CSP), a $p$-pass algorithm receives the constraints of an instance sequentially, making $p$ passes over the input in a fixed order, with the goal of approximating the maximum fraction of satisfiable constraints. We show near optimal space lower bounds for streaming CSPs, improving upon prior works. (1) Fei, Minzer and Wang (\textit{STOC 2026}) showed that for any CSP, the basic linear program defines a threshold $\alpha_{\mathrm{LP}}\in [0,1]$ such that, for any $\varepsilon > 0$, an $(\alpha_{\mathrm{LP}} - \varepsilon)$-approximation can be achieved using constant passes and polylogarithmic space, whereas achieving $(\alpha_{\mathrm{LP}} + \varepsilon)$-approximation requires $\Ome
A Lower Bound for Grothendieck's Constant
arXiv:2603.22616v1 Announce Type: cross Abstract: We show that Grothendieck's real constant $K_{G}$ satisfies $K_G\geq c+10^{-26}$, improving on the lower bound of $c=1.676956674215576\ldots$ of Davie and Reeds from 1984 and 1991, respectively.
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!