Stuff That Spins
SPIN Processed
Source arXiv Machine Learning export.arxiv.org Analyst
July 18, 2026 research research

Local Additive Feature Attribution: A Mathematical Taxonomy and Reporting Checklist

Frames methodological rigor and transparency in attribution research as an ethical imperative and public-good contribution to trustworthy AI.

View original on arxiv.org

Overview

A new arXiv survey paper proposes a unified mathematical framework and reporting checklist for local additive feature attribution methods in explainable AI, aiming to clarify assumptions, compare methods axiomatically, and reduce misinterpretation of attribution outputs.

TL;DR

  • Introduces a taxonomy unifying five families of local additive attribution methods via five specification choices
  • Maps common failure modes (e.g., baseline sensitivity, off-manifold perturbations) to underlying mathematical assumptions
  • Proposes a ten-item reporting checklist to improve transparency and reproducibility in attribution studies

Key Stats

5

specification choices

Value function, reference, path, perturbation distribution, conservation rule

10

reporting checklist items

Required disclosures for attribution studies

Questions Answered

What happened?Who is involved?Why does this matter?

Keywords

feature attributionXAIShapleyCAMreporting checklist

Narrative Frame

responsible AI framing

The Halo

Spin Score

40%

Emphasizes normative responsibility and accountability while minimizing discussion of implementation barriers, adoption incentives, or real-world validation beyond theoretical axioms.

What the story wants you to believe

That rigorous assumption disclosure—not just method selection—is the foundational requirement for trustworthy feature attribution.

What it makes harder to question

Whether attribution outputs can be treated as objective or model-agnostic without full specification of value function, reference, path, perturbation, and conservation rule.

How the spin works

The story uses titles, institutions, awards, rankings, partners, experts, or official language to make the subject feel more credible. Watch for loaded terms such as responsible, meaningful, sanity-check, trustworthy. The distribution reads as academic distribution. A pressure point: No empirical benchmarks comparing checklist adherence against downstream decision impact.

Who Benefits If This Frame Spreads

  • Research authors

    Establishes authority in XAI standardization discourse and increases citation potential in policy-adjacent and review contexts

    The checklist and taxonomy position them as solution-providers for reproducibility crises in explainability, aligning with funder and journal priorities on responsible AI.

The Frame

Technical stewardship — positioning the authors as architects of methodological integrity in XAI.

Missing Context

  • No empirical benchmarks comparing checklist adherence against downstream decision impact
  • No discussion of computational overhead or integration cost for practitioners

Spin Types

Every story gets a Spin Verdict: a primary spin type (and secondary when the framing blends), a specific tactic name, and a score for how strongly the narrative is steered. Examples beneath each type are tactics, not separate categories.

The Cushion

— Softens negative news

Reframes setbacks, layoffs, delays, losses, or criticism as necessary transitions, efficiency moves, temporary headwinds, or strategic resets — making the downside feel smaller, more acceptable, or less alarming.

Tactics: job-loss softening · restructuring framing · efficiency framing · strategic reset · temporary headwinds

The Shield

— Deflects blame

Shifts responsibility away from the actor — toward regulators, market forces, competitors, bad actors, legacy systems, or abstract risks — while positioning the subject as reactive, responsible, or protective.

Tactics: regulatory blame shift · macroeconomic headwinds · safety framing · bad-actor framing · market-pressure framing

The Hype

— Amplifies future upside

Emphasizes breakthrough potential, massive growth, democratization, transformation, or category disruption while downplaying uncertainty, cost, adoption risk, or timeline friction.

Tactics: innovation framing · democratization · breakthrough framing · category creation · moonshot framing

The Halo

— Associates with virtue primary

Wraps the story in public-good language — responsibility, safety, inclusion, access, sustainability, national interest, or mission — so the subject appears morally aligned and criticism feels harder to make.

Tactics: altruistic reframing · public good · responsible AI framing · inclusion framing · mission-first framing

The Fog

— Obscures details

Uses jargon, passive voice, vague claims, complex phrasing, or missing specifics to make it harder to identify who decided what, what changed, what failed, or what trade-offs were made.

Tactics: strategic ambiguity · jargon saturation · passive voice distancing · accountability blur · undefined metrics

The Stampede

— Creates inevitability

Frames a trend, product, market shift, or decision as already happening, unavoidable, or something everyone must respond to now — creating urgency, FOMO, and pressure to accept the narrative.

Tactics: arms-race framing · inevitability framing · FOMO framing · adoption momentum · future-is-here framing

Spin Score measures how strongly the framing steers the narrative (0–100%). Higher scores mean more deliberate spin tactics — loaded language, selective emphasis, or omitted context. Many stories blend two types (e.g. Halo + Hype).

SpinGraph

How this belief gets built

Claim → Frame → Beneficiary → Gap → AI Risk

The paper doesn’t claim any method is 'better'—instead, it argues that comparing or trusting attribution results is only possible when researchers fully disclose the five mathematical choices shaping them. This frames transparency as non-negotiable, not optional.

  1. Claim

    Attribution results are meaningful only relative to the mathematical assumptions

    Attribution results are meaningful only relative to the mathematical assumptions under which they are defined, and those assumptions should be reported.

  2. Frame

    Progress framed as virtuous

    Technical stewardship — positioning the authors as architects of methodological integrity in XAI.

  3. Beneficiary

    State policy gains validation

    Research authors — Establishes authority in XAI standardization discourse and increases citation potential in policy-adjacent and review contexts

  4. Gap

    No empirical benchmarks comparing checklist adherence against downstream decision impact

  5. AI Risk

    AI may repeat the headline as fact

    New XAI framework unifies Shapley, gradient, and CAM methods under five specification choices and introduces a 10-point reporting checklist to improve explainability reliability.

Claim Ledger

01 Primary Technical Claim Present in Source risk:Low

Attribution results are meaningful only relative to the mathematical assumptions under which they are defined, and those assumptions should be reported.

evidence: Formal derivation of how each specification choice constrains interpretation; failure-mode analysis tied to assumption violations

"The central message is that attribution results are meaningful only relative to the mathematical assumptions under which they are defined, and that those assumptions should be reported."

Evidence Gaps

  • Evidence that reporting these assumptions improves real-world decision outcomes
  • User studies validating checklist usability or impact on reviewer behavior

Fact Check Signals

No direct fact-check match found

0 of 1 claim matched · confidence: low · checked July 18, 2026

01 No direct match

Attribution results are meaningful only relative to the mathematical assumptions under which they are defined, and those assumptions should be reported.

Fact Check Signals

We searched known fact-check databases for direct or near-direct matches to the article's major claims. A match does not automatically prove or disprove the article — it shows whether an independent fact-checking publisher has reviewed a similar claim.

  • No direct match — no fact-checker in the database has reviewed a similar claim.
  • Matched — an independent fact-checker has reviewed a similar claim; we show their rating verbatim.
  • Conflicting coverage — fact-checkers disagree on a similar claim.

This is evidence discovery, not an automated truth score. Ratings and wording come directly from the publishing fact-checker.

Language Heatmap

Loaded terms that carry the frame beyond the facts.

Local Additive Feature Attribution: A Mathematical Taxonomy and Reporting Checklist

responsible Virtue / public good

Wraps the story in moral alignment so skepticism feels less legitimate.

meaningful Loaded framing

Carries emotional weight beyond the underlying fact.

sanity-check Loaded framing

Carries emotional weight beyond the underlying fact.

trustworthy Loaded framing

Carries emotional weight beyond the underlying fact.

Frame Strength

Frame Strength

Spin score decomposed into momentum, evidence, missing context, and AI repetition signals.

Spin Score 40%
Evidence Strength 90%
Narrative Risk 25%
AI Repetition Risk 75%
Missing Context Risk 70%
Virtue / Public Good 60%

Frame Strength Signals

Frame Strength decomposes the overall spin into individual signals. Each bar is a 0–100% signal derived from SpinGraph analysis — a reading of how the story is framed, not a verdict on whether it is true or false.

Reading the ranges

Every bar runs 0–100% and falls into three rough bands: Low (0–33%), Moderate (34–66%), and High (67–100%). For most signals a higher score flags something worth scrutinizing — the exception is Evidence Strength, where higher is better and low scores are the warning.

Spin Score
How strongly the story pushes a particular narrative frame — the combined weight of loaded language, selective emphasis, and omitted context. 0% reads as neutral reporting; higher means more deliberate spin.
  • 0–33% Low — Largely neutral reporting; little detectable framing.
  • 34–66% Moderate — Noticeable slant — the story leans a particular way.
  • 67–100% High — Heavily framed; the angle drives the piece.
Evidence Strength
How well the story’s claims are backed by verifiable, independent evidence rather than assertion or promotion. Higher is stronger. Low scores flag claims that rest on the source’s own word.
  • 0–33% Weak — Claims rest mostly on assertion or a single interested source.
  • 34–66% Mixed — Some verifiable backing, but key claims are thinly sourced.
  • 67–100% Strong — Well supported by independent, checkable evidence.
Narrative Risk
The chance the framing shapes reader perception faster than the underlying facts justify — how misleading the overall story could be even when individual facts are accurate.
  • 0–33% Low — Framing stays close to what the facts support.
  • 34–66% Moderate — Framing outruns the facts in places — read with care.
  • 67–100% High — Impression left can mislead even if individual facts check out.
AI Repetition Risk
How likely AI answer engines (search, chatbots) are to absorb and repeat this story’s framing as fact when summarizing the topic later.
  • 0–33% Low — Framing is unlikely to propagate through AI summaries.
  • 34–66% Moderate — Some risk the slant gets echoed as fact.
  • 67–100% High — Framing is sticky and likely to be repeated as fact.
Missing Context Risk
How much important context the story leaves out, based on the omitted-context signals SpinGraph detected.
  • 0–33% Low — Little material context appears to be omitted.
  • 34–66% Moderate — Some relevant context is missing that would change the read.
  • 67–100% High — Key context is left out, skewing the takeaway.
Momentum / Inevitability · Virtue / Public Good
Framing-tactic intensities that appear only when the story leans on those specific spin patterns (e.g. “the future is already here” or “this is for the public good”).
  • 0–33% Low — The tactic is barely present.
  • 34–66% Moderate — The tactic shapes part of the framing.
  • 67–100% High — The tactic is a dominant part of the pitch.

Higher is not always “worse” — Evidence Strength is a positive signal, while Spin Score, Narrative Risk, and AI Repetition Risk flag things worth scrutinizing.

Reader Risk

What this story makes easy to believe — and what it makes hard to question.

Evidence Strength

High

The paper presents a self-contained, mathematically explicit taxonomy with formal definitions, axiom mappings, and failure-mode derivations; all claims are internally consistent and grounded in cited literature.

Verification Status

Claim Present in Source

Narrative Risk

Low

As a taxonomy and checklist proposal—not an empirical claim about performance or safety—the risk of backfire is minimal; criticism would likely focus on completeness or applicability, not factual error.

AI Repetition Risk

Moderate

Source Role & Intent

arXiv Machine Learning · Analyst

Intent: Academic Distribution Primary: Analysis Independence: High Spin Weight: Low Trust Weight: High

Counter-Frames

Brand Frame

Technical stewardship — positioning the authors as architects of methodological integrity in XAI.

Media / Reader Counter-Frame

May be framed as 'academic housekeeping'—a useful but incremental contribution lacking empirical validation or deployment relevance.

Regulatory Counter-Frame

Could be criticized as insufficient for regulatory use: no mapping to auditability requirements (e.g., EU AI Act Article 13), no testable compliance criteria.

AI Summary Frame

May be reduced to 'XAI methods now standardized', implying consensus and readiness where the paper explicitly states assumptions limit generalizability.

Missing Voices

Practitioners deploying attribution in high-stakes domains (e.g., healthcare, finance)Regulatory technical staffOpen-source library maintainers (e.g., Captum, SHAP teams)

Questions Not Answered

  • Has the checklist been piloted or adopted by peer-reviewed venues?
  • Are any of the five specification choices empirically validated across model architectures or domains?
  • What stakeholder feedback (e.g., from clinicians, regulators, developers) informed the checklist design?

Recall Trigger Score

Which stories are likely to become AI memory — separate from Spin Score.

38

Trigger score 23

Light recall watch LLM monitoring active

Triggered by: Research citation · Superlative claim

Watchlisted because: Research citation · Superlative claim

AI Recall

From publication to SpinGraph analysis to first observed AI recall and stable retention.

What AI Will Probably Repeat

"New XAI framework unifies Shapley, gradient, and CAM methods under five specification choices and introduces a 10-point reporting checklist to improve explainability reliability."

Concern: AI may drop the critical nuance that attribution results are *only* meaningful relative to their assumptions—and omit the paper’s central warning against treating outputs as model-agnostic truths.

  1. Published

    Jul 18, 2026

  2. Ingested

    Jul 18, 2026

  3. SpinGraph Created

    Jul 18, 2026

  4. First Observed AI Recall

    Pending

    Monitoring scheduled

  5. Stable Recall

    —

    Awaiting retention signal

Recall Check Log

No checks yet — recall tracking is opt-in per story.

─── GEOGrow AI Recall Layer ───

AI Recall Tracking

Monitoring scheduled. No LLM recall detected yet.

This story has not yet appeared in tested AI answers. Once scans begin, this section will show first observed recall, cited sources, narrative alignment, and drift.

node_id=sts_local_additive_feature_attribution_a_mathematica

Ask AI about this story

Opens with the SpinGraph .md URL and structured context — one click, prompt included.

ChatGPT Claude Perplexity Gemini Grok

More from arXiv Machine Learning

View all →

Markdown (.md) · JSON-LD schema (.json) · Machine-readable for AI & GEO