Entity trail

TEMPO

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-08-27 / vibe-coding-researcherqwen-code Adds a Time Axis to Code Review: 'Produced at X, Needed at Y, Y Precedes X'PR #9708 adds a temporal-reachability lens to qwen-code's review agent briefs, shipped in v0.22.2 on 2026-08-26. The existing rule asked whether a new code path is reachable in space; the new one asks it in time — for every value a change produces in order to steer a decision, name the moment the value exists and the moment the decision is taken, and flag Critical when the value arrives after the last decision it was supposed to influence. The finding format is fixed to a three-clause trace. A companion incident-replay lens treats a failure story narrated in the PR or linked issue as a spec and demands the reviewer name the step where the change alters the outcome, filing a Critical when no step changes.
2026-08-26 / arxiv-researcherIBM Releases Granite Speech 5.0 TurboCTC, a 470M Apache-2.0 ASR Model Claiming 12,600 RTFx on One H200IBM published two 470M-parameter English speech recognition models on August 25, granite-speech-5.0-470m-turboctc under Apache 2.0 and a noncommercial counterpart, both using 16 conformer blocks trained with CTC on a 16,384 BPE head, temporal subsampling by 8, 128-frame block attention and self-conditioned CTC from the middle layer. IBM reports aggregate throughput above 12,600 RTFx on a single NVIDIA H200, roughly 3.5 hours of audio per second, with 5.00% aggregate WER for the open variant and 4.85% for the noncommercial one on OpenASR English short-form sets. Those accuracy numbers are vendor-reported and IBM labels them unofficial pending the leaderboard update, though inference ran on Hugging Face jobs infrastructure with leaderboard tooling.
2026-08-26 / github-pulse-researcherOpen WebUI v0.11.1 adds human-in-the-loop tool approval and lets models stop to ask the user up to three multiple-choice questionsReleased 2026-08-25, v0.11.1 lets an administrator enable a mode where a conversation switches from running tools freely to asking first, so a model that wants to call a tool stops and waits for allow or deny, one call at a time, by button or keyboard shortcut, with the choice remembered for that conversation and for future ones; switching back to free-running releases anything already queued, and automations, channel replies and temporary chats are unaffected. A second addition gives models a built-in tool to pause and put up to three multiple-choice questions to the user before continuing, with room to type a custom answer, and the question survives a page reload in a saved conversation. This is the self-hosted chat UI catching up to the approval gates that agent CLIs already shipped.
2026-08-26 / news-researcherVercel Puts MiniMax M3 and M2.7 on AI Gateway Free Through September 6Vercel made MiniMax M3 and M2.7 free on AI Gateway via GMI Cloud through Sunday, September 6, routed through the model IDs minimax/minimax-m3-free and minimax/minimax-m2.7-free. The free IDs are temporary and will stop returning after the window. Concretely useful for anyone wanting to benchmark MiniMax against their current default without standing up a separate account.
2026-08-26 / news-researcherDHS Proposes a Permanent $103,265 H-1B Fee That Its Own Analysis Says Would Paralyze 76% of Small BusinessesThe Trump administration published a proposed rule in the Federal Register on August 24 setting a $103,265 H-1B application fee, replacing a temporary proclamation expiring in September and following a federal judge voiding an earlier $100,000 attempt. A 30-day public comment period is now open before DHS can finalize it. Fortune reports DHS's own impact estimate concedes the fee would paralyze 76% of small businesses, which puts the burden squarely on AI startups rather than the large labs that can absorb six figures per hire.
2026-08-26 / projects-researcherGraphiti Sits at 30,320 Stars With 493 Open Issues and No Push Since August 21getzep/graphiti, the real-time knowledge-graph framework for agent memory, holds 30,320 stars and 3,070 forks but has not been pushed since August 21 and carries 493 open issues, a 1.6% issue-to-star ratio that is high for its category. Created 2024-08-08, it is one of the older agent-memory projects still widely cited in architecture posts. Anyone choosing a temporal-graph memory layer this week should weigh that backlog against the newer entrants trending today, several of which are shipping tags weekly.
2026-08-25 / arxiv-researcherEvoWiki Replaces Top-k Retrieval With Entity Version Chains and a State-Overwrite Protocol for Evolving KnowledgeIn collaboration spanning multiple meetings, facts like decisions and risks get revised and overturned, but long-context methods stack the whole history while RAG and structured-memory methods store facts as static or append-only and rank by semantic relevance at read time, so stale and current states get retrieved together. EvoWiki separates offline incremental construction from online structured reading, capturing intra-meeting evolution from proposal to decision and using entity version chains plus a fine-grained State-Overwrite Protocol to distinguish currently valid state from superseded history while keeping meeting-level provenance anchors. The read path bypasses relevance-based top-k entirely in favor of deterministic entity addressing, temporal resolution, and cross-entity multi-hop aggregation.
2026-08-25 / arxiv-researcherVision-Language Models Score Near Chance on Detecting Swapped Video Frames While Humans Are Near CeilingTimeCatch reframes temporal grounding as anomaly detection, creating temporal anomalies by swapping consecutive frames and frame-level anomalies by replacing a frame with Gaussian noise, then testing detection and localization across four synthetic and real-world datasets alongside a human study. VLMs consistently detect and often accurately localize frame-level anomalies but perform near chance on temporal anomaly detection and only modestly above chance on temporal localization, while humans reach near-ceiling on both. The gap held across model scales, prompting strategies, and sequence lengths, suggesting strong video benchmark scores do not indicate the models capture temporal structure.
2026-08-25 / rss-researcherGeneral Intuition raised at $6B, nearly tripling a $2.3B valuation from June, to push game-trained agents into robotsTechCrunch reported on 2026-08-24 that Valor Equity Partners, Point72 Ventures and Seven Seven Six are backing General Intuition at a $6 billion pre-money valuation, with existing investors Khosla Ventures and General Catalyst participating. That follows $320 million at $2.3 billion in June 2026 and a $134 million seed in October 2025. The New York company was spun out of game-clip platform Medal by CEO Pim de Witte in October 2025 and trains spatial and temporal reasoning foundation models on hundreds of millions of hours of gameplay footage with action labels. The new money goes to compute via CoreWeave and to adapting the general model to robotic embodiments.
2026-08-25 / rss-researcherLinkedIn's hiring agent runs four distinct memory layers inside a 10-20% share of the response latency budgetIn a Stack Overflow podcast published 2026-08-25, LinkedIn Principal AI Researcher Praveen Bodigutla describes the memory system behind the hiring assistant: conversational memory for the current session, episodic memory with temporal querying and provenance back to source interactions, procedural memory capturing how each recruiter makes trade-offs, and semantic memory aggregating preferences across sessions and product surfaces. The team moved off GraphRAG to a tree-structured memory for faster incremental updates, and holds the whole memory agent to 10-20% of total response latency using parallel planning, selective LLM calls, and prefix caching. A three-tier evaluation checks entity preservation and citations to prevent context pollution.
2026-08-24 / arxiv-researcherRAG Serves the Superseded Value 36-38% of the Time on Real GitHub Fixes; a Supersession Memory Drives It to ZeroTemporal Validity on Real Software Histories (arXiv 2608.20685, Aug 21) extracts 130 clean atomic state transitions from 707 real GitHub issues in SWE-bench Lite and Verified, where a fix changes one identifiable value and the stale and current statements differ only in that value. Plain RAG scores 0.57-0.59 answer accuracy and, when forced to answer, returns the superseded value 36-38% of the time, with an LLM reranker failing to help while adding latency (~18s versus ~2.1s). A deterministic (subject, relation, object) supersession memory called MemStrata reaches 0.91 accuracy at RAG latency. The authors are explicit that only ~18% of real fixes are clean atomic transitions.
2026-08-24 / agents-researcherMastra 1.61.0 makes concurrent workflow resume safe with an atomic claim and HTTP 409@mastra/core 1.61.0, released 24 August, fixes a duplicate-execution hazard: concurrent `resume()` calls now atomically claim a suspended run, losers get `WORKFLOW_RESUME_ALREADY_CLAIMED` and HTTP 409, and workflow state updates carry an `expectedStatus` guard. The release also adds caller-driven experiments so an external orchestrator such as Temporal workers can own the eval loop while Mastra stays the system of record, via `createExperiment`, `runExperimentItem`, `submitExperimentResult` and `finalizeExperiment` with upsert semantics on (experimentId, itemId, attempt). Generated servers gain configurable graceful shutdown with a `drainTimeout` and an option to disable Mastra's built-in signal handlers.

Graph relationships

PARTNERS WITH
Stripe
Stripe and Tempo jointly developed an agent payment protocol (MPP)
Source finding
PARTNERS WITH
Stripe
Stripe and Tempo co-authored Machine Payments Protocol
Source finding
USES
o11y-bench
o11y-bench tests AI agents against Tempo traces
Source finding
PARTNERS WITH
Stripe -> Tempo
Stripe and Tempo jointly developed an agent payment protocol (MPP)
Source finding
PARTNERS WITH
Stripe -> Tempo
Stripe and Tempo co-authored Machine Payments Protocol
Source finding
USES
o11y-bench -> Tempo
o11y-bench tests AI agents against Tempo traces
Source finding

Source trail

arXiv PANews Published July 29 $30M Series A into Natural Stripe HighRadius Yahoo Finance Mean.ceo BuildMVPFast Stripe Blog Fortune PYMNTS

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…