arXiv AI
Jul 21
Constrained Path Reasoning: Measuring When Committed Stages Earn Their Cost
arXiv:2607. 17240v1 Announce Type: new Abstract: When does a committed intermediate stage in an LLM reasoning pipeline earn its cost?
Certified Speculative Execution for Untrusted AI Agents
arXiv:2606. 31023v1 Announce Type: cross Abstract: Hard-constrained sequential decision systems have no certified way to spend the test-time compute of modern AI: executing the multi-step drafts of a learned policy or a frozen LLM forfeits the feasibility guarantee a trusted solver provides, while invoking the solver at every step forfeits the speed the AI offers.
Related stories
arXiv Machine Learning
Aug 11
RouteGuard: Certifying Routing Gain in LLM Multi-Agent Systems When Complementarity Is Not Enough
arXiv:2608. 07583v1 Announce Type: cross Abstract: Multi-agent LLM systems route among model-backed advisors, yet a deployer rarely knows before shipping whether routing will help at all.
arXiv AI
Jun 11
Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents
arXiv:2606. 11688v1 Announce Type: cross Abstract: Long-horizon LLM agents are not trusted to run unattended: with no human watching, they confidently report success they never verified.
arXiv AI
Jun 19
Think Again or Think Longer? Selective Verification for Budget-Aware Reasoning
arXiv:2606. 19808v1 Announce Type: new Abstract: Test-time reasoning is increasingly used as a serving-time control knob, but extra reasoning is not uniformly valuable: it can repair failed attempts, waste compute on already-correct answers, or introduce harmful answer changes.
arXiv AI
Jun 9
Cost-Aware Speculative Execution for LLM-Agent Workflows: An Integrated Five-Dimension Method
arXiv:2606. 07846v1 Announce Type: cross Abstract: LLM-agent workflows chain model calls and tool invocations, and spend most of their wall-clock time waiting on upstream operations before downstream ones can start.
arXiv AI
Jul 22
Falsifiable Release Gates for Self-Improving Systems: Standing Invariants at Scale
arXiv:2607. 13070v2 Announce Type: replace-cross Abstract: Safety claims for self-improving agent runtimes are almost always self-graded: a policy file, a guardrail, a promise in a README.
arXiv AI
Jul 2
Self-Evolving Agents with Anytime-Valid Certificates
arXiv:2607. 00871v1 Announce Type: new Abstract: Self-evolving agents violate the assumption behind most learning-theoretic guarantees: the data, evaluator, components, and hypothesis space are produced by the policy being updated.
arXiv AI
Aug 6
The Order Is the Guarantee: Verifier-Budgeted Code Deletion with Static-First Learned Proposals
arXiv:2608. 04611v1 Announce Type: cross Abstract: Frontier coding models now match or exceed strong human reference points on programming benchmarks, yet benchmark success does not imply maintainable software.
arXiv AI
Jun 30
Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds
arXiv:2606. 29654v1 Announce Type: new Abstract: Multi-agent deliberation among LLMs can improve reasoning, but deployment requires deciding when the current answer is reliable enough to act on and when it should be escalated to human review.
arXiv AI
Jun 8
Hierarchical Certified Semantic Commitment for Byzantine-Resilient LLM-Agent Collaboration
arXiv:2606. 07316v1 Announce Type: cross Abstract: Byzantine collaboration among large-language-model agents requires a finality-control primitive: given delivered stochastic, structured natural-language proposals, the protocol must decide whether the round supports a commit, what kind of commit, or a typed safe abort.
arXiv AI
Jul 28
A Frozen 12B Beats Frontier Models on Verified Work: 100% Accuracy, 0 Tokens, Bit-Exact, Forever
arXiv:2607. 23806v1 Announce Type: cross Abstract: Improving a language model today means retraining it: enormous compute, a new opaque model each cycle, non-deterministic output.
arXiv AI
Aug 5
Approximate Speculative Decoding
arXiv:2608. 03447v1 Announce Type: cross Abstract: Speculative decoding accelerates autoregressive generation by verifying a draft block with a target model in parallel.