Papers for
distributed systems developers
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
Symbolic method speeds probabilistic timing models verification hundredfold
Symbolic Abstraction Refinement
Abstract: Autonomous systems choose actions whose outcomes are uncertain and are subject to time constraints. Probabilistic timed automata (PTAs) model exactly this, and verifying one answers a question about the system's decision-making: the maximum probability of reaching a goal, or the minimum probability of reaching a hazard, that a controller can guarantee. Verification normally works over symbolic states, each pairing a discrete location with a region of clock values, and its difficulty is that the best behavior generally varies within such a region. The PRISM verifier's most capable PTA method responds by splitting regions until each is uniform, rebuilding the transitions that reach them and propagating the split backward through the state graph of a stochastic game, a cascade of graph reconstruction that partitions regions the property never depends on. We present SYMAR, which instead makes the value function symbolic: each symbolic state carries a function from sub-regions of its clock region to probability bounds, so refinement is the appearance of new sub-regions in that function, and previously generated states are never split nor their incoming transitions reconstructed. Both sources of PTA nondeterminism, the timing and the choice of actions, are captured implicitly by the preimage computation that produces those sub-regions. Heuristic forward search then bounds how much refinement is done at all, expanding an And-Or graph from the initial state along the most promising behaviors so that effort is confined to the property-relevant fragment, guided by bounds and preferred actions from a single unrefined run of the existing method; a partial-refinement backup keeps the otherwise combinatorial sub-region count lean, and every stored interval provably contains the true probability. On the PRISM PTA benchmarks, whose subproblems recur in autonomous systems (distributed agreement under contention, leader election, and self-configuration without central coordination), SYMAR is up to 108x faster, solves an instance on which the existing method exhausts the time limit, and is markedly less sensitive to the deadline constant, resolving one instance in a 37-node search graph, unchanged as the deadline triples, where the existing method needs 86,380 abstract states and 73 refinement rounds.
Safety problems in networks with shared counters become solvable under restrictions
Safety Analysis of Parameterized Networks with Shared Counters
Abstract: We study networks with an unbounded number of processes sharing some natural counters that can be decremented, incremented and tested to zero. We look at two safety problems: the first one asks whether there is a network with an execution bringing a process in a given final state and the other one asks the execution to bring all the processes in the final state. In their full generality, these two problems are undecidable, but we show that decidability can be regained by either using a single counter or by forbidding the use of zero-tests (as it is is the case in Vector Addition Systems with States). For these decidable cases, we obtain low complexity results, the problems being solvable either in non-deterministic logarithmic space or in polynomial time.
Generalized sorting matches ordinary sorting efficiency with new algorithm
A Near-Optimal Generalized Sorting Algorithm
Abstract: We continue the study of the classic \emph{generalized sorting problem}, a basic variant of ordinary sorting. There are $n$ elements and an (initially hidden) total ordering $π$ over them. Unlike in ordinary sorting, however, only \emph{some} comparisons are allowed: those given by the edges of an undirected graph $G = (V, E)$ on the elements. How many comparisons are needed to find the total ordering? In ordinary sorting, $Θ(n \log n)$ comparisons are necessary and sufficient. For generalized sorting, however, the best known upper bound has been $\tilde{O}(n^{3/2})$ since the work of Huang, Kannan, and Khanna (FOCS 2011), who even asked whether the problem might in fact require $n^{1+Ω(1)}$ comparisons. We show that it does not: we give a randomized algorithm that uses $O(n \log^4 n)$ comparisons in expectation. Thus, up to logarithmic factors, generalized sorting is no harder than ordinary sorting.
New algorithms reduce congestion in connecting network graphs
Sublinear Approximations for Spanning Tree Congestion
Abstract: The spanning tree congestion of a connected graph \(G\) is the minimum, over all spanning trees \(T\) of \(G\), of the maximum number of edges of \(G\) crossing a cut induced by deleting an edge of \(T\). In the spanning tree congestion problem we are asked to compute a spanning tree with minimum congestion. We describe two sublinear-ratio approximation algorithms for the spanning tree congestion problem (STC): a simple $\Ot(n^{8/9})$-approximation, and a more involved $\Ot(n^{2/3})$-approximation, where the notation $\Ot$ hides factors polynomial in $\log n$.. Until now, no algorithm with sublinear ratio was known. We also prove that for any fixed $0<\varepsilon<1$, approximation of the spanning tree congestion within $2-\varepsilon$ is NP-hard. This improves the previous best known lower bound $6/5$.
Faster updates for dynamic graph maximal independent sets against adaptive changes
Faster Dynamic Maximal Independent Set against Adaptive Adversaries
Abstract: We study the fully dynamic maximal independent set (MIS) problem: maintaining an MIS in an undirected graph on $n$ vertices with at most $m$ edges under edge insertions and deletions. Against an oblivious adversary, which fixes the edge update sequence in advance, the randomized algorithms of Behnezhad et al. (2019) and Chechik and Zhang (2019) achieve polylogarithmic update time. Against an adaptive adversary, which may choose each update after observing the previously reported MISs, the best previously known update bound was $O(m^{2/3})$, given by the deterministic algorithm of Gupta and Khan (2018, 2021). Assuming the combinatorial Boolean matrix multiplication hypothesis, Bernstein et al. (2026) proved that any combinatorial algorithm against an adaptive adversary requires at least $m^{1/2-o(1)}$ update time even under edge insertions alone, leaving a polynomial gap between upper and lower bounds. We present a randomized fully dynamic MIS algorithm with $\widetilde O(m^{5/8})=\widetilde O(m^{2/3-1/24})$ update time against adaptive adversaries. This gives the first polynomial improvement over the $m^{2/3}$ bound for adaptive fully dynamic MIS.
Compact routing methods improve path efficiency with low memory use
Optimal (Randomized) Compact Routing Schemes
Abstract: Compact routing schemes route messages along short paths while every vertex stores little information. We study the trade-off between the stretch and the storage of compact routing schemes in undirected graphs, in four models: labeled and name-independent schemes, with average or worst-case local storage. We improve the best known stretch in all four models. Our main result is a labeled routing scheme with expected stretch $2k-1$ and $\widetilde O(n^{1/k})$ average local storage, in which only the routing is randomized. Under the Erdős girth conjecture, no scheme with this storage has stretch below $2k-1$, so our scheme is optimal, up to polylogarithmic factors in the storage. With deterministic routing, we obtain stretch $\approx2.47k$, the best known deterministic stretch with average storage. Worst-case local storage is the classical setting of compact routing. The leading coefficient $4$ in the stretch $4k-5$ of Thorup and Zwick~[SPAA'01] was improved only by Chechik~[PODC'13], to $\approx3.68$ in unweighted graphs. In unweighted graphs, we obtain expected stretch $(\frac83k+O(1),O(k))$, and deterministic stretch $\approx3.05k$. Name-independent routing is harder, since the source knows only the name of the destination. The best known stretch with $\widetilde O(n^{1/k})$ local storage in unweighted graphs was $96k$, by Abraham, Gavoille, and Malkhi~[DISC'04]. We improve it by more than a factor of $8$. With average storage, we obtain expected stretch $4k-5$, which matches the labeled scheme of Thorup and Zwick. With worst-case storage in unweighted graphs, we obtain expected stretch $(8k+O(1),O(k))$. All our schemes for unweighted graphs extend to weighted graphs with a small loss in the stretch, and still improve the previous unweighted bounds.
Verusweave benchmark tests fixing proofs in concurrent rust code
VerusWeave: A Benchmark for Proof Repair in Concurrent Verus Programs
Abstract: Concurrent software underpins critical infrastructure, including operating systems and databases, yet thread interactions can have bugs that bypass conventional testing. Verus extends Rust with formal verification capabilities, enabling developers to specify, implement and prove the correctness of concurrent code in the same language. Although coding agents have been explored for automated proof construction, existing Verus benchmarks offer limited coverage of concurrency-specific examples that connect atomic operations to abstract protocols. To address this gap, we introduce VerusWeave, the first benchmark dedicated to concurrent proof repair, derived from four open-source projects. It comprises 75 focused tasks and 17 full-module variants that exercise features used to verify low-level concurrent code. Tasks remove selected proof annotations and require agents to fill gaps, while preserving executable code and specifications. The evaluation harness validates repairs using Verus and a change checker, recording solutions, execution traces, and resource usage. An initial evaluation of four general-purpose coding-agent harnesses achieved 75-85% success on focused tasks and 29-47% on full-module variants, highlighting the challenge of autonomous module-wide repairs. VerusWeave provides a reusable foundation for evaluating automated proof construction for concurrent software.
Approximating minimum spanning trees despite corrupted distances
Approximating Minimum Spanning Trees in Metrics under Semi-Random Corruptions
Abstract: We study the problem of finding minimum spanning trees in semi-random corrupted metrics. In this problem proposed by Bateni, Dharangutte, Jayaram, and Wang (COLT'24), we start with a metric space $\mathcal{M} = (X, d)$. The adversary independently corrupts each pairwise distance with probability $p$. The algorithm only observes these possibly corrupted distances and must output a spanning tree whose expected cost (in the true metric) is multiplicatively close to the minimum spanning tree cost. Prior work showed that losing a factor of $Ω(\log n)$ is necessary, even if the algorithm may adaptively select $O(n/\log n)$ points and probe the true distances between them. However, no upper bounds were known. We study three regimes based on whether the adversary can only increase, decrease, or arbitrarily change the corrupted distances. We show the following: (1) In the increase-only setting, we give a polynomial-time $O(1)$-approximation algorithm for any constant $0 \le p < 1$. (2) In the decrease-only setting, we give an $O(\log n)$-approximation algorithm for any constant $0 \le p < 1$. Additionally, we show an $O(\log^4 n)$-approximation algorithm that runs in quasi-polynomial time $n^{O(\log n)}$, for any constant $p \leq 0.1$. Moreover, we show that a polynomial-time algorithm computing an $n^{1/3-\varepsilon}$-approximation is unlikely to exist. (3) In the general setting, where the adversary can both increase and decrease distances, we show an $O(\log n)$-approximation algorithm for every $0 \leq p < \frac{1}{4}$. Further, we show that no $o(n)$-approximation is possible for $\frac{1}{2} \leq p < 1$. Our algorithms answer the main open question in Bateni et al.'s work; moreover, the $O(\log n)$-approximations in the latter two settings are information-theoretically optimal.
How group interactions shape spreading dynamics on complex networks
Spreading Dynamics on Higher-Order Networks: A Structure-Mechanism-Behavior Perspective with Applications
Abstract: Spreading in real-world networks, including modern communication and networked systems, often unfolds through group interactions rather than isolated pairwise contacts. Higher-order network representations such as simplicial complexes, hypergraphs, and their temporal, multilayer, and directed extensions provide a principled way to encode group interactions. Unfortunately, existing studies on spreading over these substrates remain fragmented and largely model-centric, offering limited insight into how higher-order representations and contagion mechanisms jointly shape emergent dynamics. This review addresses this gap by introducing a structure-mechanism-behavior three-level perspective that provides a systematic mapping from higher-order network structures and spreading mechanisms to collective behaviors. Within this framework, we compare how different higher-order network representations encode group interactions, introduce a taxonomy of contagion mechanisms, and analyze their structural carrying capacity, defined as the ability of a given representation to support specific contagion mechanisms and the associated dynamical behaviors. We further map specific structure-mechanism combinations to three hallmark emergent behaviors, thereby offering a unified cross-level perspective beyond existing model-centric surveys. These insights are particularly relevant for communication systems involving group interactions, such as social information spreading, cooperative protocols, and distributed networked processes. Lastly, leveraging this three-level perspective, we offer a roadmap for guiding and controlling dynamics on higher-order networks, while discussing future directions in representation choice, mechanism design, and data-driven inference.
Vertex fault tolerant spanners with improved size efficiency
New Fault-Tolerant $(α,β)$-Spanners
Abstract: We present new constructions of vertex fault-tolerant (VFT) $(α,β)$-spanners with substantially improved size bounds. Given a bound $f$ on the number of vertex faults, an $f$-VFT $(α,β)$-spanner is a subgraph of $G$ that preserves an $(α,β)$-spanner of $G- F$ for every fault set $F \subseteq V$ with $|F| \le f$. Most existing constructions of VFT $(α,β)$-spanners rely on a black-box sampling technique that incurs an exponential blow-up compared to their non-fault-tolerant counterparts. We develop two independent techniques for constructing $f$-VFT $(α,β)$-spanners that improve the dependence on $f$, reducing it to polynomial overhead. For every unweighted graph $G=(V,E)$ on $n$ vertices, stretch parameter $k$, and fault parameter $f$, we show: (i) $f$-VFT $(O_ε(k^ε), O_ε(k))$-spanners with $\tilde{O}(f^{exp(1/ε)}\cdot n^{1+1/k})$ edges for any $ε\in (0,1)$. For any fixed constant $ε$, this introduces a $poly(f)$ overhead compared to the non-faulty constructions of [Ben-Levy and Parter, SODA 2020] and [Chechik and Lifshitz, SODA 2026]. (ii) For any $δ\in (0,1)$, an $f$-VFT $\left(O_δ(\sqrt{k}),O_δ(k) \right)$-spanner of size $\tilde{O}(f^{1+δ} \cdot n^{1+\frac{1}{k}})$. Hence, it incurs a size overhead of only $f^{1+δ}$ compared to the above-mentioned non-faulty analogs. (iii) $f$-VFT $(k,k-1)$ spanner of size $O(k^3 \cdot f \cdot n^{1+\frac{1}{k}})$, nearly matching the lower bound $Ω(f^{1-\frac{1}{k}}n^{1+\frac{1}{k}})$ known for $f$-VFT $(2k-1,0)$ spanners by [Bodwin et al., SODA 2018]. The first result is obtained via a new adaptation of the Baswana--Sen framework to the VFT $(α,β)$ setting. The other two bounds follow from a highly involved analysis of a natural extension of the fault-tolerant greedy spanner.
Intelligence extends itself by creating new meanings and purposes
Co-Recursive Semantic Self-Extension: A Mathematical Foundation for Open-Ended Intelligence
Abstract: Artificial intelligence has advanced through scale, data, and computation, yet remains bounded by its representational framework. We ask what makes intelligence open-ended. Every inferential system selects meaningful distinctions, limiting what it can learn. Within a fixed purpose, performance reaches a ceiling set by its semantic domain. Teleological succession, in which purposes arise from earlier ones, faces a similar limit: Boolean-closed purpose-transformations exhaust their reachable semantic territory. We shift from purposes to inferential systems. Drawing on Kant and Gödel, we model cognitive states as partial evaluations over a formal language. Unresolved propositions may be undecided or undecidable. Reasoning can resolve the former but not the latter, leaving a frontier for extension. Non-trivial reductions of undecidable propositions expose undecidable intermediate premises whose reflective reification supports multiple decisive extensions. Finite trace-guided generation remains fallible, requiring meta-level selection and polycontextural exploration. Using Kolmogorov complexity, we show that no architecture with a fixed state description, transition mechanism, and initial condition can generate histories incorporating irreducible algorithmic information. Because physical systems are fixed in this sense, we examine collectives. Coupling finitely many finite agents does not remove the bound. Extension requires persistent external memory preserving semantic content and historical coordinates. Collective generative capacity depends on balancing overlap for transfer with difference for generation. We propose semantic self-extension as a design principle: intelligence should not merely resolve ambiguity within existing meanings, but transform unresolved structures into new spaces of meaning, generating new distinctions, questions, representations, and purposes.
Deterministic methods speed up information spread in radio networks
Amplified Pruning and Path Selectors for Deterministic Radio Gossiping
Abstract: We study deterministic gossiping in unknown strongly connected directed radio networks with $n$ vertices, $m$ arcs, and distinct identifiers in $\{1,\ldots,N\}$, where $N=n^{\mathcal O(1)}$. Messages are unrestricted; collision detection is unavailable. Our protocol combines amplified reception certificates, adaptive pruning, and permutation selectors. Each pruning iteration costs $\mathcal O(n\log^2n\log\log n)$ rounds. A two-step quadratic contraction limits each receiver and each class of identical closed in-neighborhoods to $\mathcal O(\log\log n)$ qualifying selections. Under an explicit degree-concentration condition, our protocol takes $\mathcal O(m\log^2n(\log\log n)^2)$ rounds. For directed regular graphs with common in- and out-degree $Δ$, our strategy achieves $\widetilde{\mathcal O}(\min\{n\sqrtΔ,n^2/Δ\})$, extending the identifier range of Wu and Chrobak and improving their high-degree bound. At completion entry, let $a\ge1$ count original sources not yet credited as globally disseminated. Permutation selectors reduce our balanced completion budget by a factor of $\log^{2/3}(a+1)$ relative to interleaved path selectors. Our geometric protocol attains $\widetilde{\mathcal O}(k^2+nτ+\min\{u^2,(nu)^{2/3}\})$ for an original path cover of scale $k$, weight $τ$, and $u$ exceptions. Every graph also admits $\widetilde{\mathcal O}(\min\{n^{4/3},N+(mn)^{3/5}\})$ rounds. Only $n,N$ are inputs. Within the stated graph classes and identifier ranges, these results answer the polynomial-identifier and average-degree questions of Wu and Chrobak. The unrestricted general-graph cases remain open.
Parallel greedy matching keeps vertex connections stable in random graphs
Degree Stability of Parallel Greedy Matching in Random Graphs
Abstract: We study a natural synchronous greedy matching process in Erdős-Rényi random graphs $\mathbb{G}(n,p)$ and give a rigorous analysis of this parallel stochastic dynamics. The process evolves as follows: in each round, every unmatched vertex proposes to a uniformly random neighbour, and mutually proposing pairs are matched and removed from the graph, after which the process repeats on the residual graph. We consider the density regime $p=Θ(\log^{1+ε} n/n)$ for any constant $ε>0$, which lies just above the threshold for the existence of a perfect matching. We show that, with high probability, for $Θ(\log^{1+ε} n)$ rounds, every unmatched vertex keeps a constant fraction of its initial degree in the residual graph, simultaneously for all vertices and all rounds. By then, the process has matched a constant fraction of the vertices. Our analysis tracks fine-grained invariants that control the evolution of individual vertex degrees, rather than directly tracking the global degree distribution. At the local level, we analyze the per-round degree loss of a vertex via a neighbourhood-exposure martingale. We show that structural sparsity of $\mathbb{G}(n,p)$ yields $O(1)$ variance and thus the per-round degree loss is strongly concentrated around its conditional expectation. At the global level, these local concentration bounds control the increments of a martingale tracking cumulative degree loss over many rounds, allowing us to show that vertex degrees remain stable for $Ω(\log^{1+ε} n)$ rounds.
Fast Tendermint agreement bug exposed and two fixes proposed
Revisiting Fast Tendermint: An Agreement Violation and Two Repairs
Abstract: Fast Tendermint reduces Tendermint's good-case latency to two communication steps for n>=5f+1. We show that its published round-change rule violates Agreement: one Byzantine process suffices for every f>=1, and for f>=2 two Byzantine processes suffice without double signing. The failure occurs when a process advances past a round without recording the value that may have been decided there. The Quint specification that accompanies the protocol does not allow this kind of round skipping and is not affected; the violation lies in the published pseudocode. We present two repairs. Walk requires a process to leave each round on a quorum of that round's precommits and to precommit before leaving. It keeps the recorded valid state and removes the wait after a precommit quorum. Justify carries the evidence needed to accept a proposal in the proposal itself, removing both the recorded valid state and the proposer's synchronization wait. Both satisfy Agreement and Validity under asynchrony and Termination under the partially synchronous gossip model of Tendermint and Fast Tendermint. Both retain proposal timeouts. The main text presents the executions that violate Agreement and explains the two repairs. The appendices provide proofs, a brief comparison with other round-change designs, and an extension to chains.
Network design method boosts connectivity using density graphs and learning
Toward Maximizing Algebraic Connectivity via Density-Dependent Backbone Construction and Reinforcement Learning Completion
Abstract: This paper studies an open network design problem of constructing graph topologies with maximum algebraic connectivity under fixed numbers of vertices and edges. A density-dependent approach is proposed that decomposes the design task into two stages: a high-quality, low-cost structural backbone construction and a reinforcement learning completion stage. The method selects the backbone based on density, using complete multipartite graphs in dense regimes and Cayley graphs in low-to-intermediate regimes, and then refines the remaining edges through a learning-based procedure. Numerical results show that the proposed method achieves high algebraic connectivity with favorable runtime scalability, outperforming existing methods. These results support a structural-family-first strategy for scalable network design.
Robots meet reliably using fixed distance despite disoriented views
Rendezvous under Variable Disorientation:The Algorithmic Power of Fixed Unit Distance
Abstract: We study deterministic Rendezvous of two robots with lights under weak geometric and synchronization assumptions. Our focus is on Variable Disorientation (VD), where each robot may arbitrarily rotate, reflect, and rescale its local coordinate system at every \Look, and on VD+Fixed-Unit-Distance (FUD), where each robot's local unit distance remains fixed. We consider the one-sided visibility models FSTA and FCOM, and schedulers ranging from the energy-restricted RSYNCH and R-RSYNCH to fully asynchronous executions. Under VD, we prove color-independent impossibility results for FSTA under SSYNCH and for FCOM under ASYNCH, and establish sharp two-color boundaries under RSYNCH and R-RSYNCH. Under VD+FUD, we introduce periodic distance classes, which turn multiplicative changes of the physical distance into predictable cyclic phase shifts. This technique yields a two-color self-stabilizing non-$L$-Rendezvous algorithm for both FSTA and FCOM under RSYNCH and R-RSYNCH even with Non-Rigid movement. Under Rigid movement, periodic distance classes further yield an optimal three-color non-quasi-self-stabilizing algorithm for FSTA under SSYNCH, a four-color FSTA algorithm under $M$-atomic ASYNCH, and a four-color FCOM algorithm under CM-atomic ASYNCH. By guarding the latter phase mechanism with two additional colors and replacing its initial swap by passive initialization, we obtain a six-color FCOM algorithm under full ASYNCH, improving the previous twelve-color upper bound by a factor of two. These results show that a fixed local metric can serve as persistent cyclic memory, and clarify how metric stability, one-sided visibility, and scheduler atomicity jointly determine Rendezvous solvability and color complexity.
Translating bach programs for better verification of shared data
Modelling Shared-Space Coordination in mCRL2: a Bach-to-mCRL2 Translation Framework
Abstract: Although significant research has focused on the theory and implementation of data-based coordination languages, the critical aspect of their automated verification using model-checking techniques remains underexplored, which is essential for ensuring reliability and correctness in distributed systems. Existing tools, such as Anemone, provide a solid foundation for reachability-based verification of Bach programs. While they effectively analyze properties expressed in terms of state attainability, extending support to more expressive temporal specifications-such as liveness properties or invariants over shared space contents-remains an open opportunity. Such extensions are important to capture comprehensive system behaviors, for instance, ensuring that "a request is always matched by a response" or that "no message is silently lost". Addressing this limitation, we propose an automated translation from Bach to mCRL2, which explicitly represents the shared space, thereby enabling the use of mCRL2's mu-calculus model checker to verify complex properties beyond simple reachability. Complementing this translation, we introduce a systematic method to analyze the shared space in mCRL2 concerning data reachability, by directly expressing properties over the shared space's contents in the mu-calculus, thus providing a clearer framework for verification. This approach enables a novel verification process that combines action-based properties with state-based properties over the shared space contents, a largely unexplored area in current coordination-language verification approaches, offering a new dimension of analysis
Cobalt speeds up distributed large model training by reducing data transfer
Cobalt: Leveraging Expert Co-activation for Efficient Distributed MoE Training
Abstract: Mixture-of-Experts (MoE) has increasingly become a mainstream approach for scaling large language models, as it expands model capacity while keeping computation cost nearly constant. Training large-scale MoE models relies on Expert Parallelism (EP), which distributes expert replicas across GPUs and exchanges tokens through all-to-all communication. The efficiency of EP is often constrained by two system bottlenecks: cross-node token transfers are limited by inter-node bandwidth, while skewed expert workloads lead to imbalanced computation across GPUs. Prior work mitigates these bottlenecks based on per-expert workload statistics, but overlooks the fact that experts could share the communication. In this work, we empirically present the observation that many pairs of experts are frequently co-activated by individual tokens. Motivated by this, we present Cobalt, an efficient MoE training framework that leverages expert co-activation to reduce cross-node traffic and workload imbalance. Cobalt adopts a two-stage expert layout planner that adapts expert layout to the evolving expert co-activation and workload conditions. It periodically co-locates frequently co-activated experts on the same node to reduce the cross-node communication, and performs per-step intra-node adjustment to rebalance the workloads. Subsequently, we develop a communication-aware task assignment method that routes tokens to fewer remote nodes based on the current expert layout. Experiments on 32 B200 GPUs show that Cobalt achieves up to 1.53-2.41 times (1.28-1.89 times on average) of speedup compared to existing MoE training frameworks, while reducing cross-node token traffic by 75.74%-99.26%.
Simple shortest-path trees give good low-stretch spanning trees
Low-Stretch Spanning Trees via Smoothed Analysis of Dijkstra's Algorithm
Abstract: Given an undirected weighted graph $G$, a $γ$-approximate low-stretch spanning tree (LSST) $T \subseteq G$ is a tree that approximates the distance metric of $G$ up to a $γ$-factor in expectation. Currently, existing algorithms to find a provably good LSST carefully construct an approximate shortest-path tree from an arbitrary source. The resulting algorithms are intricate. In contrast, practitioners observed that a much simpler heuristic performs surprisingly well: choose an arbitrary root, run Dijkstra's algorithm, and use the resulting shortest-path tree as an LSST. In this paper, we give a smoothed analysis of shortest-path tree algorithms, such as Dijkstra's algorithm, that explains this behavior. We show that adding a small perturbation to the weights of the input graph suffices to turn the shortest path tree rooted at an arbitrary node in the resulting graph into an $\tilde{O}(1)$-approximate LSST. We further show that the set of perturbations can be computed efficiently from few low-diameter decompositions (LDDs). Thus, our proof is also constructive in the sense of giving a novel approach to computing LSSTs.
Signature-free blockchain methods cut consensus delays significantly
Simple and Fast Signature-Free Blockchain Consensus
Abstract: Signature-free protocols avoid the cost of post-quantum signatures. We present two simple signature-free blockchain consensus protocols for eventual synchrony with optimal good-case commit latency (three message delays for \(f<n/3\) and two for \(f<n/5\)), optimistic responsiveness, a block time of only two message delays without speculation, and \(O(n^2)\) communication per view. They instantiate Generic Simplex, a blockchain consensus construction parameterized by a new abstraction called view agreement. The same construction also captures Simplex, Minimmit, and a new synchronous signature-free protocol with optimal good-case commit latency of two message delays for \(f<n/4\).
HyperParallel-FSDP speeds up training of massive AI models on Ascend SuperPods
HyperParallel-FSDP: Topology-Aware Fully Sharded Training with Layout-Driven Muon on Ascend SuperPods
Abstract: Declarative SPMD programming uses tensor sharding descriptions to drive distributed execution, separating parallelization from model code. However, the evaluated PyTorch DTensor stack dispatches every operator below autograd, incurring repeated dispatch and metadata costs, while lacking an inexpensive end-to-end validation path. Existing FSDP and distributed Muon implementations also mismatch two-tier supernode topologies: FSDP relies on explicit parameter packing and unpacking, and Muon's whole-matrix orthogonalization conflicts with parameter sharding. We observe that distributed tensors need only express sharding semantics at the tensor API boundary above autograd, allowing differentiation and kernels to operate on plain tensors. Based on this insight, we present HyperParallel-FSDP, featuring: (1) dual-mode DTensor execution, using one sharding plan for both a production mode with one-time layout resolution and no steady-state dispatch overhead, and a validation mode with end-to-end metadata propagation, fail-fast checks, and gradient-equivalence testing; (2) topology-aware FSDP, with zero-copy intra-supernode collectives, fused inter-supernode reduction, and a cross-layer backward pipeline that avoids waits on slow links; and (3) layout-driven distributed Muon, with sharding-derived communication groups, deduplicated orthogonalization, and shape-fused Newton-Schulz iterations. On Atlas 900 A3 SuperPoD, HyperParallel-FSDP scales from 16 dies to 384 cards (768 ranks), sustaining 421k tokens/s for a 505B-parameter MoE while FSDP communication uses 2.9% of step time. It reduces mean step time by 29.7% versus PyTorch FSDP2 and 25.5% versus Megatron DDP, with Pearson correlation above 0.999997 over 1,000 steps. Distributed Muon improves profiler step time by 5.4-16.0% over competing systems. Source code is available at https://atomgit.com/mindspore/hyper-parallel.
Cloud native MPI enables selective process relocation for resilient computing
XMPIaaS: Towards Cloud Native MPI via Cooperative Process Migration
Abstract: Message Passing Interface (MPI) has been the dominant programming model for High Performance Computing (HPC) for three decades, and as HPC workloads increasingly migrate to cloud infrastructure for scalability and cost efficiency, MPI applications must contend with an execution environment fundamentally unlike traditional supercomputers: ephemeral resources, dynamic pricing and preemptable instances. In such a volatile setting, the ability to relocate running MPI processes between nodes without restarting the job is a necessity for cost-effective, resilient execution. Existing approaches either require restarting the entire job from a global checkpoint, or transparently intercepting the full MPI stack at prohibitive complexity. To address these challenges, we propose \name, a cooperative migration system for MPI that enables selective process group migration on-the-fly. When a cloud instance is scheduled for preemption, only the affected ranks are relocated while the remaining processes briefly quiesce and resume in place, avoiding the cost of a full-job checkpoint. \name tackles this through a cooperative protocol between the MPI process management runtime and rank processes. We expose an \texttt{XMPI\_quiesce} interface built atop the MPI Sessions API that allows applications to mark safe migration points, and we extend the Hydra process manager to orchestrate the full migration lifecycle: rank quiescence, CRIU checkpoint/restore, proxy relaunch on the target node, and seamless rank reconnection. We evaluate and show that the cooperative quiesce phase accounts for less than 1.4\% of total migration downtime, and that this downtime is governed by the migrating node's rank count alone, independent of job size, and the instrumentation introduces no measurable overhead during normal execution.
Sublinear size coalitions can control outputs in complex systems
A Resolution of Friedgut's Conjecture on Influential Coalitions
Abstract: We prove that, for every constant $\varepsilon>0$ and every function $f:Σ^n\to\{0, 1\}$, there is a coalition of $O(n/\sqrt{\log n})$ coordinates and a target output $b\in\{0, 1\}$ such that, after the remaining coordinates are sampled uniformly and independently, the coalition can choose its values to make the output equal to $b$ with probability at least $1-\varepsilon$. The bound is independent of the alphabet size and also holds for monotone Boolean functions on $[0,1]^n$, resolving a conjecture of Friedgut (Combinatorics, Probability and Computing, 2004). Unlike the Boolean cube setting, where Kahn, Kalai, and Linial (FOCS, 1988) give a coalition bound of $O(n/\log n)$, no sublinear bound independent of the alphabet size was previously known. In collective coin flipping, our result gives the first sublinear bound on the number of bad players needed to force a fixed output with probability at least $1-\varepsilon$ in any one-round protocol with independent uniform messages, regardless of the message length. A key ingredient in our proof is an encoding that lets us relate the influence of a function on a product space to the $p$-biased influence of the encoded function. We then rely on a structure theorem of Hatami (Annals of Mathematics, 2012) for functions with small $p$-biased influence to bias the encoded function.
Faster federated learning with first order bilevel optimization
Federated stochastic bilevel optimization with fully first-order gradients
Abstract: Federated stochastic bilevel optimization has been actively studied in recent years due to its widespread applications in machine learning. However, most existing federated stochastic bilevel optimization algorithms require the computation of second-order Hessian and Jacobian matrices, which leads to longer running times in practice. To address these challenges, we propose a novel federated stochastic variance-reduced bilevel gradient descent algorithm that relies solely on first-order oracles. Specifically, our approach does not require the computation of second-order Hessian and Jacobian matrices, significantly reducing running time. Furthermore, we introduce a novel learning rate mechanism, i.e., a constant single-timescale learning rate, to coordinate the update of different variables. We also present a new strategy to establish the convergence rate of our algorithm. Finally, the extensive experimental results confirm the efficacy of our proposed algorithm.
Deterministic labeling improves fault-tolerant connectivity checks in networks
Deterministic Edge-Fault-Tolerant Connectivity Labeling Schemes with Nearly Optimal Label Size
Abstract: For an undirected graph $G = (V,E)$ and a fault bound $f$, an edge-fault-tolerant connectivity labeling scheme assigns short labels to vertices and edges, so that for any vertex pair $(s,t)$ and failed edge set $F\subseteq E$ with $|F|\leq f$, the connectivity between $s$ and $t$ in $G-F$ can be answered by inspecting only the labels of $s$, $t$ and edges in $F$. In this paper, we present a labeling scheme that uses $O(\log^{2}n)$-bit labels that can be computed in deterministic polynomial time. This improves upon the previous $\tilde{O}(\sqrt{f})$ deterministic bound of [Long, Pettie, Saranurak'25], and even slightly improves the $O(\min\{f+\log n,\log^{2}n\log f\})$ randomized bound of [Dory, Parter'21] and [Long, Pettie, Saranurak'25] when $f = Ω(\log^{2}n)$. Moreover, for a general $f$, this is the first labeling scheme that produces an $\tilde{O}(1)$-size labeling which is simultaneously correct across all queries. Our approach combines the cycle-space-based labeling scheme from Dory and Parter with a recent result by [Knauer'26] on sparse cycle bases.
Maximum mutual visibility sets found and built on cactus graphs
The Maximum Mutual Visibility Set on a Cactus Graph and the Self-stabilizing Constructions
Abstract: Given a graph $G=(V,E)$, let $S$ ($\subseteq V$) be a set of vertices. Two vertices are \emph{mutually visible} if there exists a shortest path in $G$ between them that does not contain any other vertex of $S$. A set $S$ is a \emph{Mutual Visibility Set} (\MVS) if every pair of vertices in $S$ is mutually visible. The concept of \MVS s in graphs has attracted significant attention since its introduction, as it provides an important structural property of graphs. However, determining a maximum \MVS\ in general graphs is computationally intractable; the decision problem of whether a graph admits an \MVS\ of size at least $k$ has been shown to be \emph{NP-complete}. Thus, prior work has focused on finding maximal \MVS s or restricting attention to specific graph classes. Cactus graphs form a fundamental low-treewidth class, yet the maximum \MVS\ problem for this class remains open. In this paper, we first determine the size of maximum \MVS~in cactus graphs, and introduce two self-stabilizing algorithms that construct such sets. The first algorithm uses a single BFS tree and stabilizes in $O(D)$ rounds with $O(\log n)$ bits per process on average; the second one uses parallel BFS trees and stabilizes in $O(|C_{\max}|+|T_{\max}|)$ rounds, which we show to be asymptotically tight as a function of these two parameters, even on graphs where $|C_{\max}|+|T_{\max}| = o(D)$.