Lint report · stored run
From Detection to Refusal: Safer LLMs via Circuit-Guided Weight Scaling
0errors
0warnings
0info
23claims read
2428numbers extracted
29references
How well we read this paper
3 observations
These are notes about our reading of the paper, not about the paper. They affect how much weight the findings below can carry.
Extraction warning 1
- numbers-are-identifier-fragments 41 of 2428 extracted numbers (2%) are digits inside identifiers, not measurements. e.g. '2' in 'Safety Rate (%) ↑\nLLaMA3\nLLaMA2\nMistral\n', '3' in 'Safety Rate (%) ↑\nLLaMA3\nLLaMA2\nMistral\n', '2' in 'Safety Rate (%) ↑\nLLaMA3\nLLaMA2\nMistral\n'. Rules that compare numbers across sections can report a contradiction between an identifier and a real value.
Note 2
- compile-warning references: the reference list carries no entry numbers; citation keys are enumeration order [1]..[N], not labels printed in the paper, so a key may not match the [N] used in the body
- compile-warning semantic extraction: dropped 13 items whose quote was not found in the paper
Some checks could not run
- RL002 reported nothing, and it could not reach OpenAlex — so its silence here is not evidence that this paper is clean.
- RL003 reported nothing, and it could not reach OpenAlex — so its silence here is not evidence that this paper is clean.
Everything else on this page was computed from the paper itself and is unaffected. This is a limit of this run, not a finding about the paper.
Some checks did not apply to this paper
- RL005 did not apply to this paper — it links no repository we could read, so there was no released configuration to compare against. Its absence here says nothing about whether the paper and its code agree.
Compiler extraction log (2)
references: the reference list carries no entry numbers; citation keys are enumeration order [1]..[N], not labels printed in the paper, so a key may not match the [N] used in the body semantic extraction: dropped 13 items whose quote was not found in the paper
No findings on this paper.
That is a result, not a failure. The compiler read the paper without errors, and no rule found something it could prove with evidence — this product would rather say nothing than say something vague.
Read the extraction notes above before taking this as a clean bill of health: they say which checks were working with incomplete information.
Rule notes (30)
RL001 funnel arithmetic: sentences_with_from=10 endpoint_pairs=4 claims_compared=1 emitted=0 [dropped: not-a-measurement=0]
RL001 funnel delta-column: structured_tables=18 with_exactly_one_delta_column=0 rows_checked=0 emitted=0
RL001 funnel cross-location: quant=2428 prose=1060 prose_with_metric=129 prose_with_dataset=195 prose_with_metric_and_dataset=49 distinct_buckets=2 emitted=0 [dropped: not-a-measurement=2 not-in-quote=143]
RL001 funnel identifier-drift: settings_checked=2 sections_stating_one_value=0 emitted=0
RL002 citation entailment: attempted 17 of 17 ranked cited claims (31 in-text markers from 29 bibliography entries); 5 cited works could not be identified, 7 had no retrievable text, 0 model quotes failed verbatim verification, 0 were topically unrelated to the retrieved text.
RL002 candidate ranking (why the top 17 were chosen): [21] [introduction, absolutist wording, title/sentence overlap near zero, no DOI/arXiv id]; [15] [introduction, title/sentence overlap near zero, no DOI/arXiv id]; [22] [introduction, title/sentence overlap near zero, no DOI/arXiv id]; [3] [experiment, names an artefact (PAIR), absolutist wording, title/sentence overlap near zero, no DOI/arXiv id]; [28] [experiment, names an artefact (CircuitBreakers), absolutist wording, title/sentence overlap near zero, no DOI/arXiv id]; [11] [introduction, no DOI/arXiv id]; [13] [experiment, names an artefact (WikiText), absolutist wording, title/sentence overlap near zero]; [8] [experiment, names an artefact (Malicious Instruction Dataset), title/sentence overlap near zero, no DOI/arXiv id]
RL002 label distribution over 5 classified citations: SUPPORTED=5
RL002 base-rate self-check: not tripped (0/5 = 0% NOT_SUPPORTED, threshold 40%).
RL002 attribution check (case B): 13 candidates name an artefact, 12 of them made the top-N and were checked against their cited source
RL002: 3 citations skipped because the sentence reports this paper's own results (the citation there is a comparison pointer, not evidence for the sentence)
RL002: 1 markers skipped because the claim is jointly attributed to several works
RL002: 1 enumerated-citation sentences skipped (they cite >=3 works, so the claim is about the set and cannot be refuted by any single cited work)
RL002: 5 cited works unresolvable (first 3: [{"key": "[11]", "title": "Reinforcement learning from human feedback", "why": "no work record matched the bibliography title"}, {"key": "[8]", "title": "Catastrophic jailbreak of opensource llms via exploiting generation", "why": "no work record matched the bibliography title"}, {"key": "[20]", "title": "Concept bottleneck large language models", "why": "no work record matched the bibliography ti)
RL003: 3 queries -> 128 unique works (arXiv search 77, category listings 0, scholarly graph 51 [crossref 51], 0 graph works matched to an arXiv id); source errors during retrieval: arXiv 0, graph 0
RL003: parsed the reference list of 10/12 comparable papers (2 unreadable) — that is the PeerUsage denominator
RL003: enriched 1 of 2 candidates with their own abstract (1 unavailable, 0 over budget); 0 dropped as published after this paper
RL003: scholarly graph for this run = crossref
RL003 funnel (358 candidates): already_used_by_the_paper -7, cited_only_from_related_work -66, is_the_paper_itself -1, peer_usage_below_quorum -103, survey_or_artifact -38, title_names_no_method -141 || kept: citation_context_unknown +66 || survived-cheap-gates 2 -> with-abstract 1 -> comparable 2 -> above-threshold-0.65 0
RL003 candidate: 'Direct preference optimization' score 0.432 (T 0.00 D 1.00 M 0.33 P 0.40 R 0.71; 4/10 peers cite it) -> not reported
RL003 candidate: 'Inference-time intervention' score 0.412 (T 0.00 D 1.00 M 0.33 P 0.30 R 0.71; 3/10 peers cite it; title-only, no abstract available) -> not reported
RL003: 2 comparable candidates all scored below 0.65 — nothing to report for this paper
RL004 funnel: components=7 contribution_backed=3 candidates=3 ablations_found=6 matched_by_ablation=3 unmatched=0 experiments=5
RL004 component M1 'Harmful Detection Heads': covered (candidate)
RL004 component M2 'Safety Neurons': covered (candidate)
RL004 component M4 'Circuit-Based Safety Editing (CBSE)': covered (candidate)
RL004 component M5 'attention-head scaling': covered (filtered earlier)
RL004 component M6 'safety-neuron scaling': covered (filtered earlier)
RL004 component M7 'activation patching': covered (filtered earlier)
RL004 component M8 'targeted component ablation': covered (filtered earlier)
RL004 funnel: ranked=0 emitted=0 suppressed_by_cap=0 dropped_unverified_quote=0
Reading the report as data
This page is a rendering of the report the pipeline wrote. The JSON below
is the stored artefact, unedited — the same document
research-lint --format json produces.
loading…