Token-Efficient Data Reasoning Agents via Adaptive Structuring of Unstructured Data

Hugging Face Daily Papers Papers
token-efficiency llm-agents unstructured-data adaptive-structuring speculative-structuring cost-reduction

Summary

The paper proposes agentic data cracking, a method to adaptively structure unstructured data during LLM reasoning to reduce token consumption and costs, achieving significant cost cuts while maintaining accuracy on benchmarks.

Valuable data remains embedded in unstructured sources: web pages, reports, contracts, filings, earnings calls, and PDFs. The big bet in enterprise AI is deploying LLM agents that reason over this data to answer complex questions for every knowledge worker. Agents can do this today, but at prohibitive cost. Each question repeatedly opens large documents to recover scattered evidence, consuming up to a million tokens. However, if the data were already structured, the same question would reduce to a cheap database lookup. For example, on FanOutQA benchmark, reasoning over an ideal pre-structured store is 28X cheaper, and the gap grows to orders of magnitude as questions fan out over more documents. Yet structuring everything in advance is not viable: documents hold vastly more possible structure than any workload will use, and the useful structure and documents are unknown until queries arrive. We propose agentic data cracking, a method that structures unstructured data adaptively and speculatively as a byproduct of reasoning itself. Structuring is adaptive because observed queries decide when it happens and what matters, and speculative because it goes beyond the current question. Whenever the agent opens a document to answer, a cracking sub-agent forks from the already-loaded context at marginal cost and extracts grounded structure likely to serve related future queries. Over time, an increasing share of queries is fully covered by structured data and answered without opening a document, keeping agentic accuracy at close to RAG cost. On FanOutQA, extended with merely one related question per test question, cracking cuts cost by 53% while preserving accuracy. Agentic data cracking is a first step toward next-generation data infrastructure for agentic reasoning over unstructured data: a shared substrate beneath the model where knowledge that reasoning already paid to uncover accumulates.
Original Article
View Cached Full Text

Cached at: 09/02/26, 07:47 PM

Paper page - Token-Efficient Data Reasoning Agents via Adaptive Structuring of Unstructured Data

Source: https://huggingface.co/papers/2608.31082

Abstract

Valuabledataremainsembeddedinunstructuredsources:webpages,reports,contracts,filings,earningscalls,andPDFs.ThebigbetinenterpriseAIisdeployingLLMagentsthatreasonoverthisdatatoanswercomplexquestionsforeveryknowledgeworker.Agentscandothistoday,butatprohibitivecost.Eachquestionrepeatedlyopenslargedocumentstorecoverscatteredevidence,consuminguptoamilliontokens.However,ifthedatawerealreadystructured,thesamequestionwouldreducetoacheapdatabaselookup.Forexample,onFanOutQAbenchmark,reasoningoveranidealpre-structuredstoreis28Xcheaper,andthegapgrowstoordersofmagnitudeasquestionsfanoutovermoredocuments.Yetstructuringeverythinginadvanceisnotviable:documentsholdvastlymorepossiblestructurethananyworkloadwilluse,andtheusefulstructureanddocumentsareunknownuntilqueriesarrive.Weproposeagenticdatacracking,amethodthatstructuresunstructureddataadaptivelyandspeculativelyasabyproductofreasoningitself.Structuringisadaptivebecauseobservedqueriesdecidewhenithappensandwhatmatters,andspeculativebecauseitgoesbeyondthecurrentquestion.Whenevertheagentopensadocumenttoanswer,acrackingsub-agentforksfromthealready-loadedcontextatmarginalcostandextractsgroundedstructurelikelytoserverelatedfuturequeries.Overtime,anincreasingshareofqueriesisfullycoveredbystructureddataandansweredwithoutopeningadocument,keepingagenticaccuracyatclosetoRAGcost.OnFanOutQA,extendedwithmerelyonerelatedquestionpertestquestion,crackingcutscostby53%whilepreservingaccuracy.Agenticdatacrackingisafirststeptowardnext-generationdatainfrastructureforagenticreasoningoverunstructureddata:asharedsubstratebeneaththemodelwhereknowledgethatreasoningalreadypaidtouncoveraccumulates.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2608\.31082

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.31082 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.31082 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.31082 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Adaptive Latent Agentic Reasoning

arXiv cs.CL

This paper introduces Adaptive Latent Agentic Reasoning (ALAR), a dual-mode framework for LLM agents that uses compact latent reasoning for routine turns and selectively escalates to explicit chain-of-thought for harder decisions, achieving up to 84.6% token reduction while maintaining task accuracy.

Agentic Graph Token Reasoning

arXiv cs.LG

Introduces agentic graph token reasoning, a method that recasts graph tokenization as part of an LLM's step-by-step reasoning process, allowing the model to dynamically choose graph views and granularity during inference. The method outperforms baselines across seven graph domains and transfers zero-shot to unseen domains.