{
  "schema_version": "1.1",
  "id": "archive:https://arxiv.org/abs/2608.21277v1",
  "slug": "2608-21277v1-1hbbfsc",
  "url": "https://feed7.dev/p/2608-21277v1-1hbbfsc",
  "title": "ConceptTS: LLM-Guided Concept Bottlenecks for Interpretable Multivariate Time-Series Forecasting",
  "why_included": "ConceptTS turns LLM-proposed concepts into executable labels and interpretable forecasting bottlenecks, enabling concept-level inspection and intervention.",
  "summary": "**ConceptTS** asks an LLM to propose task-relevant concepts and executable labeling rules, avoiding manual concept annotation. It organizes predictions through **3 bottlenecks** covering history, local forecast intervals, and the full horizon before a shared decoder produces the forecast.",
  "practical_implication": "For data agents, the useful pattern is to convert language-model domain knowledge into inspectable supervision rather than letting the LLM make the final prediction. Named activations can support debugging and direct concept-level interventions.",
  "agent_context": "**ConceptTS** asks an LLM to propose task-relevant concepts and executable labeling rules, avoiding manual concept annotation. It organizes predictions through **3 bottlenecks** covering history, local forecast intervals, and the full horizon before a shared decoder produces the forecast.\n\nFor data agents, the useful pattern is to convert language-model domain knowledge into inspectable supervision rather than letting the LLM make the final prediction. Named activations can support debugging and direct concept-level interventions.\n\nEvaluation is limited to the **Beijing Multi-Site Air Quality dataset**, where accuracy was described only as competitive with black-box baselines. The material does not quantify accuracy, labeling-rule errors, or transfer to other time-series domains.",
  "source": {
    "name": "arXiv",
    "url": "https://arxiv.org/abs/2608.21277v1",
    "published_at": "2026-08-21T16:35:44.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Paper",
  "layer": "context",
  "domains": [
    "data"
  ],
  "topics": [
    "context-engineering",
    "tool-use"
  ],
  "verification": {
    "status": "needs_review",
    "label": "Needs Review",
    "method": "unverified",
    "verified_at": null
  },
  "uncertainty": [
    "Evaluation is limited to the **Beijing Multi-Site Air Quality dataset**, where accuracy was described only as competitive with black-box baselines. The material does not quantify accuracy, labeling-rule errors, or transfer to other time-series domains."
  ],
  "connected_context": {
    "meaning": "ConceptTS turns LLM knowledge into an inspectable intermediate representation rather than trusting the model as the forecaster. Against prior context work, it extends the case for domain-native structure: generated concepts and executable rules become a debuggable interface to a conventional decoder. The evidence remains narrow, so rule quality and transfer—not merely concept readability—become the key validation gaps.",
    "corpus_size": 551,
    "generated_at": "2026-08-24T10:05:01.347Z",
    "connections": [
      {
        "title": "The Next Game Engine Won't Have a Manual — Arturo Nunez, Nereu",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=VBCDhRrvlYo",
        "feed7_url": "https://feed7.dev/p/the-next-game-engine-won-t-have-a-manual-arturo-nunez-nereu-1v4vd0y",
        "reason": "Both replace unconstrained generated behavior with a domain-native vocabulary that maps intent or knowledge into inspectable, reusable operations."
      },
      {
        "title": "SheetCompass: Hierarchical Relation Graphs for Agentic Spreadsheet Reasoning",
        "source_name": "arXiv",
        "source_url": "https://arxiv.org/abs/2608.14452v1",
        "feed7_url": "https://feed7.dev/p/2608-14452v1-17xm2vb",
        "reason": "SheetCompass preserves spreadsheet relations instead of flattening them; ConceptTS applies the same structured-context principle to forecasting by organizing information through named temporal concepts."
      },
      {
        "title": "An Exploratory Evaluation of LLM-Assisted Rewriting of Moderate-Complexity Financial Sentences for DisCoCat-Based Sentiment Analysis",
        "source_name": "arXiv",
        "source_url": "https://arxiv.org/abs/2608.07439v1",
        "feed7_url": "https://feed7.dev/p/2608-07439v1-1t2nkj0",
        "reason": "The rewriting study shows that an LLM-generated intermediate artifact must be evaluated through its downstream consumer; ConceptTS likewise needs concept and rule quality judged by forecast performance, not plausibility alone."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-08-21T16:35:44.000Z",
  "modified_at": "2026-08-21T16:35:44.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/2608-21277v1-1hbbfsc",
    "json": "https://feed7.dev/p/2608-21277v1-1hbbfsc.json",
    "markdown": "https://feed7.dev/p/2608-21277v1-1hbbfsc.md"
  }
}