Search arXiv⌕ Search

arXiv · 2610.30409

Cambium: Participant-Built Commitment Roots for Shared Digital Infrastructure

Abstract

Commitment services, such as timestamping, transparency logs and provenance registries, let a party prove that it bound itself to a record without publishing the record. Today an operator runs them, or a blockchain replaces the operator at the cost of network-wide agreement on every write. We present Cambium, a permissioned peer-to-peer service in which participants build each cycle's commitment root together, so that no single party decides which inputs the root includes. Fixed-size committees certify inputs and use a recursive-doubling merge, in which disjoint groups recompute every intermediate value, to spread root construction across all participants. A successful cycle costs each node $O(c\log(n/c))$ messages for committee size $c$, and each node stores its own $O(\log(n/c))$ proof per cycle, along with the $O(n)$ membership roster. A Go prototype on AWS with $c=4$, 8-256 nodes and simulated stragglers completes cycles in 12-148 ms and verifies proofs in under 1.5 ms, and demonstrates logarithmic growth. We show that conflicting certificates identify their signers, that a proof binds a signed input to its root, and that two conflicting roots both become final only if at least a third of the nodes are Byzantine, except with probability exponentially small in the number of committees. We also set out the membership, proof-availability and exclusion questions a community deploying Cambium must answer.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Mansoor "Manny" Ahmed, Dann Toliver, Ekaterina Pavlova. 2026-10-21. Cambium: Participant-Built Commitment Roots for Shared Digital Infrastructure. https://arxiv.org/abs/2610.30409

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Fast Tendermint: Speeding Up a Foundational Consensus Protocol

Tendermint is among the most widely studied and deployed Byzantine fault-tolerant (BFT) consensus protocols, owing in part to its native leader-rotation mechanism that subsumes complex view changes. Like most partially-synchronous BFT protocols, Tendermint tolerates $f < n/3$ Byzantine processes and decides in three communication steps. Motivated by the push for lower-latency blockchains, a recent line of work shows that consensus can be solved in two communication steps when $f < n/5$. We present Fast Tendermint, an adaptation of Tendermint to the $n > 5f$ setting that decides in two communication steps in the good case, while preserving Tendermint's leader-rotation structure. Fast Tendermint collapses Tendermint's prevote and precommit steps into a single voting step and merges the $locked$ and $valid$ state. We give proofs of agreement, validity, and termination, and a formal specification in Quint, a modern surface syntax for TLA+, used to model-check the protocol.

cs.DC↗

ATTUNE: Agentic Throughput Tuning for LLM Training

The throughput of an LLM training job depends on its recipe: how the model is split across GPUs, how memory is traded for compute, and which kernels and communication overlaps it uses. The best recipe changes with every model, accelerator and framework version. Automated tuners search a fixed space of knobs and check only that a job runs, so they miss levers outside that space and can ship a configuration that silently changes training. ATTUNE is a coding agent that tunes any working recipe unattended within a GPU-hour budget. Guided by a skill of search rules, scripts and lessons from past tuning campaigns, it diagnoses each run and changes one lever at a time. A correctness gate outside the agent refuses any argument known to change the training objective. ATTUNE tunes Amazon's production post-training recipes on H100, H200 and B200 clusters, raising throughput by 1.7x on average. It beats the published throughput of four of NVIDIA's own benchmark recipes on NVIDIA's AutoModel framework by 6-17%. On models held out from its development, it improves the strongest public recipes most where they were written for another setting, by up to 2.6x on MiniMax-M2, and least where the vendor tuned them for the same hardware. Given the same GPU-hour budget, it outperforms existing automated tuners, including Bayesian optimization, random search and OptiCo, a recent LLM-agent tuner. Its gain on GLM-4.5 is more than four times the best alternative's. The same coding agent without ATTUNE's skill gains less and fails half its launches. We release ATTUNE at https://github.com/amazon-science/attune and OpenRecipes, a leaderboard of tuned and verified training recipes, at https://openrecipes.ai.

cs.DC↗

Parallel Sparse Outputs for Tensor Compilers

Sparse tensor compilers are commonly used high-performance tools that optimize tensor algebra programs over notoriously complex sparse data structures. Despite the ever-increasing size and scale of sparse datasets, there exists astonishingly little research on developing a parallel sparse tensor compiler. Existing approaches fail to parallelize or make assumptions about access patterns or output formats that limit generality. This paper presents Birdseed, the first tensor compiler to support all loop orderings over any standard sparse format in parallel. More specifically, Birdseed supports parallel sparse scatters and directly writes into list, bytemap, hash, and dense levels of fibertree sparse formats. Our approach exhibits asymptotic memory and time improvements over all previous work. We additionally demonstrate empirical speedups on challenging kernels to parallelize, such as histogram construction and SpMSpV.

cs.DC↗