Upload folder using huggingface_hub

README.md

@@ -67,6 +67,7 @@ uv run python scripts/hydrate_defextra.py \
 
 - [`docs/defextra_hydration.md`](docs/defextra_hydration.md) (technical details, CLI flags, markers).
 - [`docs/get_pdfs.md`](docs/get_pdfs.md) (how to find PDFs).
+- [`docs/mismatch_examples.md`](docs/mismatch_examples.md) (real examples of mismatch types and fixes).
 - See [`docs/defextra_hydration.md`](docs/defextra_hydration.md) for technical details and [`docs/get_pdfs.md`](docs/get_pdfs.md) for PDF sources.
 
 ## Expected minor mismatches

defextra_required_pdfs.md

@@ -61,16 +61,16 @@
 - c1e92f1be2387d14dcfaa5e1640a9939724a312a — TITLE: An empirical examination of echo chambers in US climate policy networks (https://www.semanticscholar.org/paper/c1e92f1be2387d14dcfaa5e1640a9939724a312a)
 - c84a169e6df175c4662012d3ba7dbf8fa1b5abc9 — ‘Fake news’ is the invention of a liar: How false information circulates within the hybrid news system (https://www.semanticscholar.org/paper/c84a169e6df175c4662012d3ba7dbf8fa1b5abc9, https://doi.org/10.1177/0011392119837536)
 - daw084 — Just a subtle difference? Findings from a systematic review on definitions of nutrition literacy and food literacy (https://doi.org/10.1093/heapro/daw084)
-- doi:10.1145/3677092 — 
+- doi:10.1145/3677092 —
 - dx.doi.org/https://doi.org/10.1016/j.ipm.2021.102505 —  (https://doi.org/dx.doi.org/https://doi.org/10.1016/j.ipm.2021.102505)
 - eb29476dd81aefedf2896db42f039f003a0ec5bf — Organic or Local? Investigating Consumer Preference for Fresh Produce Using a Choice Experiment with Real Economic Incentives (https://www.semanticscholar.org/paper/eb29476dd81aefedf2896db42f039f003a0ec5bf)
 - frai-06-1225093 — Rationalization for explainable NLP: a survey
 - https://aclanthology.org/2021.findings-emnlp.101 —  (https://aclanthology.org/2021.findings-emnlp.101)
 - https://aclanthology.org/2024.lrec-main.952 —  (https://aclanthology.org/2024.lrec-main.952)
-- https://arxiv.org/abs/2312.16148 — 
-- https://link.springer.com/article/10.1007/s00799-018-0261-y — 
-- https://media-bias-research.org/wp-content/uploads/2024/07/Preprint_ICWSM_25_NewsUnfold — 
-- https://www.sciencedirect.com/science/article/pii/S0957417423021437 — 
+- https://arxiv.org/abs/2312.16148 —
+- https://link.springer.com/article/10.1007/s00799-018-0261-y —
+- https://media-bias-research.org/wp-content/uploads/2024/07/Preprint_ICWSM_25_NewsUnfold —
+- https://www.sciencedirect.com/science/article/pii/S0957417423021437 —
 - icomputing.0124 — A Survey of Task Planning with Large Language Models (https://doi.org/10.34133/icomputing.0124)
 - s10462-022-10338-7 — A survey on narrative extraction from textual data (https://doi.org/10.1007/s10462-022-10338-7)
 - s10816-016-9274-2 — Quality Assurance in Archaeological Survey (https://doi.org/10.1007/s10816-016-9274-2)

docs/defextra_hydration.md

@@ -127,3 +127,5 @@ uv run python scripts/hydrate_defextra.py \
 
 - Small mismatches are expected due to PDF/GROBID text normalization.
 - Missing exact TEI spans do **not** block hydration; hash/anchor markers are used as fallback.
+- Exact TEI spans are validated against stored hashes; if they do not match, citation‑stripped hash/anchor matching is used instead.
+- See [`docs/mismatch_examples.md`](mismatch_examples.md) for concrete examples and fixes.

docs/mismatch_examples.md

@@ -0,0 +1,90 @@
+# Mismatch classes (hydration vs reference)
+
+This page summarizes common difference types seen in the latest hydrated run.
+Examples are **short fragments** with surrounding text removed.
+
+- Total differences: 144
+- wording_change: 74
+- punctuation_only: 50
+- casing: 8
+- hyphenation: 5
+- digit_letter_spacing: 3
+- truncation: 2
+- citation_spacing: 1
+- header_or_boilerplate: 1
+
+---
+## Wording / lexical differences
+
+Small wording changes (e.g., singular/plural) or tokenization artifacts.
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| 016f7a076ac272db106fbcea056752c7307f676a | specification error | context | …Conducted mainly in the late 1970s and 1980s, wave 3 of researchwitnessed yet another round o… | …Conducted mainly in the late 1970 s and 1980 s, wave 3 of research witnessed yet another round… |
+| 0209e602acaeab882fee84e244caf574cf345ef9 | ratio bias/numerosity bias | context | …ault. In the classic ratio bias task derived from Piaget and Inhelder (1951/1975), participants are offered a prize if th… | …ault. In the classic ratio bias task derived from Piaget andInhelder (1951/1975), participants are offered a prize if th… |
+| 033b21cf1c6d3bdae587e673452b994443bf3546 | narrative | context | …he bare minimum Simply put, narrative isthe representation of a… | …bare minimum Simply put, narrative is the representation of … |
+
+## Punctuation-only differences
+
+Only punctuation differs (e.g., comma, quote style).
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| 0f7eda998bbce003745ff2fdbcaa1d9a8119368b | echo chamber | context | …ces of news, then we impose on ourselves a narrowed and selfreinforcing epistemic filter, which leaves out contrary view… | …ces of news, then we impose on ourselves a narrowed and self-reinforcing epistemic filter, which leaves out contrary view… |
+| 17734113f254a64b3bae312713edba3b1e34fb56 | Post-truth Era | context | …Indeed, today we live in what some have called a “post-truth” era, which is characterized by digital disinform… | …Indeed, today we live in what some have called a "post-truth" era, which is characterized by digital disinform… |
+| 1901.00596v4 | network embedding | definition | …ional vector representation of a node which preserves a node’s topological information… | …ional vector representation of a node which preserves a node's topological information… |
+
+## Casing differences
+
+Same text except for upper/lower case.
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| 033b21cf1c6d3bdae587e673452b994443bf3546 | narrative | definition | …The representation of an event or a series of events.… | …the representation of an event or a series of events.… |
+| 05bfced33d92944b7a0672490c371342d28ee076 | observational bias | definition | …Observed data differs systematically from the unobserved data… | …observed data differs systematically from the unobserved data… |
+| 1538c4777271ae6abb542801dac01423f4d566ad | publication bias | definition | …Significant results are more likely to be published while non… | …significant results are more likely to be published while non… |
+
+## Hyphenation / line-break joins
+
+Hyphenation caused by line breaks is joined differently.
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| 3584741 | computational narrative representations | context | …cused review, we will focus exclusively on event-based narra- tive representations. Thus, we define computational narrativ… | …cused review, we will focus exclusively on event-based narrative representations. Thus, we define computational narrativ… |
+| 3584741 | metro maps method | context | …ro maps [98, 99] are an extension of the Connect the Dots ap- proach that represents more than a single storyline using a … | …ro maps [98, 99] are an extension of the Connect the Dots approach that represents more than a single storyline using a … |
+| 3584741 | open source intelligence (OSINT) | context | …]. Although OSINT data sources leverage more than just tradi- tional news articles [38], OSINT could still benefit from ne… | …]. Although OSINT data sources leverage more than just traditional news articles [38], OSINT could still benefit from ne… |
+
+## Letter–digit spacing
+
+Spacing between digits/letters differs (`bias4` vs `bias 4`).
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| 0090023afc66cd2741568599057f4e82b566137c | omitted variable bias | context | …Omitted Variable Bias. Omitted variable bias4 occurs when one or more important variables are left out o… | …Omitted Variable Bias. Omitted variable bias 4 occurs when one or more important variables are left out o… |
+| 016f7a076ac272db106fbcea056752c7307f676a | selection bias | context | …cts of race on sentencing. Conducted mainly in the late 1970s and 1980s, wave 3 of research witnessed yet another round … | …cts of race on sentencing. Conducted mainly in the late 1970 s and 1980 s, wave 3 of research witnessed yet another round… |
+| 235c4f33d5bfc81bfa09a2458fcc0e42ef4454dc | propaganda | context | …It published workbooks and held seminars in the early 1930s aimed at promoting the ideal of "self-determination," rega… | …It published workbooks and held seminars in the early 1930 s aimed at promoting the ideal of "self-determination," rega… |
+
+## Truncation (missing tail)
+
+Hydrated text is missing the end of the reference span.
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| 016f7a076ac272db106fbcea056752c7307f676a | specification error | definition | …the omission of explanatory variable… | …the omission of explanatory variables… |
+| 0d23df558a30492946059c017343a431dc3dc172 | inter-media agenda-setting | definition | …ed theory to explain how content transfers between news medi… | …ed theory to explain how content transfers between news media… |
+
+## Citation spacing/formatting differences
+
+Citation formatting differs (e.g., `[155, 164]` vs `[155,164]`).
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| frai-06-1225093 | Explainability | context | …ndent, and AI embraces a wide variety of tasks (Miller, 2019a). We treat explainability as a specialization of interpret… | …ndent, and AI embraces a wide variety of tasks (Miller, 2019 a). We treat explainability as a specialization of interpret… |
+
+## Header/boilerplate inserted
+
+Hydrated text includes header/boilerplate not in reference.
+
+| Paper | Concept | Field | Reference (fragment) | Hydrated (fragment) |
+| --- | --- | --- | --- | --- |
+| https://arxiv.org/abs/2312.16148 | spin bias | context | …ndencies between words and phrases must be considered [110]. Spin Bias describes a form of bias introduced either by lea… | …ndencies between words and phrases must be considered [110]. Manuscript submitted to ACM The Media Bias Taxonomy Spin Bias describes a form of bias introduced either by lea… |

scripts/build_defextra_test_pdfs.py

@@ -63,7 +63,11 @@ def _build_pdf_index(
                         index.setdefault(stripped, path)
                         index.setdefault(stripped.lower(), path)
                 if stem.endswith("_fixed") or stem.endswith("-fixed"):
-                    base = stem[: -len("_fixed")] if stem.endswith("_fixed") else stem[: -len("-fixed")]
+                    base = (
+                        stem[: -len("_fixed")]
+                        if stem.endswith("_fixed")
+                        else stem[: -len("-fixed")]
+                    )
                     if base:
                         index[base] = path
                         index[base.lower()] = path

scripts/defextra_markers.py

@@ -257,6 +257,62 @@ HYPHEN_CHARS = {
 }
 SOFT_HYPHEN = "\u00ad"
 
+CITATION_BRACKET_RE = re.compile(r"\[[^\]]{0,120}\]")
+CITATION_PAREN_RE = re.compile(r"\([^\)]{0,120}\)")
+
+
+def _looks_like_bracket_citation(text: str) -> bool:
+    return any(ch.isdigit() for ch in text)
+
+
+def _looks_like_paren_citation(text: str) -> bool:
+    if not any(ch.isdigit() for ch in text):
+        return False
+    lowered = text.lower()
+    if "et al" in lowered:
+        return True
+    if re.search(r"\b(19|20)\d{2}\b", text):
+        return True
+    return False
+
+
+def strip_citations(
+    text: str,
+    *,
+    strip_brackets: bool = True,
+    strip_parens: bool = False,
+) -> str:
+    if not text:
+        return text
+    spans: list[tuple[int, int]] = []
+    if strip_brackets:
+        for match in CITATION_BRACKET_RE.finditer(text):
+            if _looks_like_bracket_citation(match.group(0)):
+                spans.append((match.start(), match.end()))
+    if strip_parens:
+        for match in CITATION_PAREN_RE.finditer(text):
+            if _looks_like_paren_citation(match.group(0)):
+                spans.append((match.start(), match.end()))
+    if not spans:
+        return text
+    spans.sort()
+    merged: list[tuple[int, int]] = []
+    for start, end in spans:
+        if not merged or start > merged[-1][1]:
+            merged.append((start, end))
+        else:
+            merged[-1] = (merged[-1][0], max(merged[-1][1], end))
+    parts = []
+    cursor = 0
+    for start, end in merged:
+        if cursor < start:
+            parts.append(text[cursor:start])
+        parts.append(" ")
+        cursor = end
+    if cursor < len(text):
+        parts.append(text[cursor:])
+    return "".join(parts)
+
 
 def tokenize_text(
     text: str,

scripts/hydrate_defextra.py

@@ -20,9 +20,11 @@ try:
         doi_suffix,
         extract_ids_from_tei,
         extract_text_from_pdf,
+        hash_token_sequence,
         normalize_arxiv,
         normalize_doi,
         normalize_paper_id,
+        strip_citations,
         tokenize_text,
     )
     from scripts.defextra_pdf_aliases import candidate_pdf_aliases
@@ -40,9 +42,11 @@ except ModuleNotFoundError as exc:
         doi_suffix,
         extract_ids_from_tei,
         extract_text_from_pdf,
+        hash_token_sequence,
         normalize_arxiv,
         normalize_doi,
         normalize_paper_id,
+        strip_citations,
         tokenize_text,
     )
     from scripts.defextra_pdf_aliases import candidate_pdf_aliases
@@ -152,8 +156,9 @@ def _cleanup_spacing(text: str) -> str:
     if not text:
         return text
     value = text
-    value = value.replace("“", "\"").replace("”", "\"")
+    value = value.replace("“", '"').replace("”", '"')
     value = value.replace("’", "'").replace("‘", "'")
+
     def _dash_repl(match: re.Match[str]) -> str:
         run = match.group(0)
         return "--" if len(run) >= 2 else "-"
@@ -287,6 +292,34 @@ def _find_pdf_hash_span(
     return None
 
 
+def _bool_flag(value: str) -> bool:
+    return (value or "").strip().lower() == "true"
+
+
+def _strip_flags(row: dict, prefix: str) -> tuple[bool, bool]:
+    keep_bracket = _bool_flag(row.get(f"{prefix}_has_bracket_citation", ""))
+    keep_paren = _bool_flag(row.get(f"{prefix}_has_paren_citation", ""))
+    return (not keep_bracket), (not keep_paren)
+
+
+def _span_matches_hash(row: dict, text: str, prefix: str) -> bool:
+    if not text:
+        return False
+    expected_hash = row.get(f"{prefix}_hash64") or ""
+    expected_sha = row.get(f"{prefix}_sha256") or ""
+    if not expected_hash or not expected_sha:
+        return False
+    strip_brackets, strip_parens = _strip_flags(row, prefix)
+    check_text = strip_citations(
+        text,
+        strip_brackets=strip_brackets,
+        strip_parens=strip_parens,
+    )
+    tokens, _ = tokenize_text(check_text, return_spans=True)
+    hash64, sha, _ = hash_token_sequence(tokens)
+    return str(hash64) == str(expected_hash) and sha == expected_sha
+
+
 def _candidate_ids(paper_id: str, doi: str, arxiv: str) -> list[str]:
     candidates = [
         paper_id,
@@ -387,7 +420,11 @@ def _build_pdf_index(pdf_dir: Path) -> Dict[str, Path]:
                     index.setdefault(stripped, path)
                     index.setdefault(normalize_paper_id(stripped), path)
             if stem.endswith("_fixed") or stem.endswith("-fixed"):
-                base = stem[: -len("_fixed")] if stem.endswith("_fixed") else stem[: -len("-fixed")]
+                base = (
+                    stem[: -len("_fixed")]
+                    if stem.endswith("_fixed")
+                    else stem[: -len("-fixed")]
+                )
                 if base:
                     index[base] = path
                     index[normalize_paper_id(base)] = path
@@ -780,8 +817,31 @@ def main() -> None:
     token_cache: Dict[str, Optional[TokenIndex]] = {}
     tei_path_cache: Dict[str, Optional[Path]] = {}
     pdf_token_cache: Dict[Path, TokenIndex] = {}
+    pdf_token_cache_stripped: Dict[tuple[Path, bool, bool], TokenIndex] = {}
+    tei_token_cache_stripped: Dict[tuple[Path, bool, bool], TokenIndex] = {}
     pdf_failed: set[Path] = set()
 
+    def _get_stripped_index(
+        cache: Dict[tuple[Path, bool, bool], TokenIndex],
+        source_path: Optional[Path],
+        source_text: str,
+        strip_brackets: bool,
+        strip_parens: bool,
+    ) -> Optional[TokenIndex]:
+        if source_path is None:
+            return None
+        if not strip_brackets and not strip_parens:
+            return None
+        key = (source_path, strip_brackets, strip_parens)
+        if key not in cache:
+            stripped = strip_citations(
+                source_text,
+                strip_brackets=strip_brackets,
+                strip_parens=strip_parens,
+            )
+            cache[key] = TokenIndex.from_text(stripped)
+        return cache[key]
+
     with args.legal_csv.open("r", encoding="utf-8", newline="") as handle:
         reader = csv.DictReader(handle)
         legal_rows = list(reader)
@@ -981,6 +1041,11 @@ def main() -> None:
             def_end = row.get("definition_char_end") or ""
             ctx_start = row.get("context_char_start") or ""
             ctx_end = row.get("context_char_end") or ""
+            def_strip_brackets, def_strip_parens = _strip_flags(
+                row,
+                "definition",
+            )
+            ctx_strip_brackets, ctx_strip_parens = _strip_flags(row, "context")
 
             if not definition and pdf_token_index:
                 span = _find_pdf_hash_span(row, pdf_token_index, "definition")
@@ -991,6 +1056,27 @@ def main() -> None:
                         span[1],
                     )
                     hydrated_from_pdf += 1
+                if not definition:
+                    stripped_index = _get_stripped_index(
+                        pdf_token_cache_stripped,
+                        pdf_path,
+                        pdf_token_index.doc_text,
+                        def_strip_brackets,
+                        def_strip_parens,
+                    )
+                    if stripped_index is not None:
+                        span = _find_pdf_hash_span(
+                            row,
+                            stripped_index,
+                            "definition",
+                        )
+                        if span:
+                            definition = _extract_with_trailing_punct(
+                                stripped_index.doc_text,
+                                span[0],
+                                span[1],
+                            )
+                            hydrated_from_pdf += 1
             if not definition and tei_token_index:
                 spec = _select_hash_specs(row, "definition")
                 if spec:
@@ -1001,6 +1087,22 @@ def main() -> None:
                             span[0],
                             span[1],
                         )
+                    if not definition:
+                        stripped_index = _get_stripped_index(
+                            tei_token_cache_stripped,
+                            tei_path,
+                            doc_index.doc_text,
+                            def_strip_brackets,
+                            def_strip_parens,
+                        )
+                        if stripped_index is not None:
+                            span = stripped_index.find_span_by_hash(*spec)
+                            if span:
+                                definition = _extract_with_trailing_punct(
+                                    stripped_index.doc_text,
+                                    span[0],
+                                    span[1],
+                                )
                 if not definition and not (def_start and def_end):
                     head_specs = _select_anchor_spec_list(
                         row,
@@ -1130,11 +1232,13 @@ def main() -> None:
                             span[1],
                         )
             if not definition and def_start and def_end:
-                definition = _extract_with_trailing_punct(
+                candidate = _extract_with_trailing_punct(
                     doc_index.doc_text,
                     int(def_start),
                     int(def_end),
                 )
+                if _span_matches_hash(row, candidate, "definition"):
+                    definition = candidate
 
             if not context and pdf_token_index:
                 span = _find_pdf_hash_span(row, pdf_token_index, "context")
@@ -1145,6 +1249,27 @@ def main() -> None:
                         span[1],
                     )
                     hydrated_from_pdf += 1
+                if not context:
+                    stripped_index = _get_stripped_index(
+                        pdf_token_cache_stripped,
+                        pdf_path,
+                        pdf_token_index.doc_text,
+                        ctx_strip_brackets,
+                        ctx_strip_parens,
+                    )
+                    if stripped_index is not None:
+                        span = _find_pdf_hash_span(
+                            row,
+                            stripped_index,
+                            "context",
+                        )
+                        if span:
+                            context = _extract_with_trailing_punct(
+                                stripped_index.doc_text,
+                                span[0],
+                                span[1],
+                            )
+                            hydrated_from_pdf += 1
 
             if not context and tei_token_index:
                 spec = _select_hash_specs(row, "context")
@@ -1156,6 +1281,22 @@ def main() -> None:
                             span[0],
                             span[1],
                         )
+                    if not context:
+                        stripped_index = _get_stripped_index(
+                            tei_token_cache_stripped,
+                            tei_path,
+                            doc_index.doc_text,
+                            ctx_strip_brackets,
+                            ctx_strip_parens,
+                        )
+                        if stripped_index is not None:
+                            span = stripped_index.find_span_by_hash(*spec)
+                            if span:
+                                context = _extract_with_trailing_punct(
+                                    stripped_index.doc_text,
+                                    span[0],
+                                    span[1],
+                                )
                 if not context and not (ctx_start and ctx_end):
                     head_specs = _select_anchor_spec_list(
                         row,
@@ -1285,11 +1426,13 @@ def main() -> None:
                             span[1],
                         )
             if not context and ctx_start and ctx_end:
-                context = _extract_with_trailing_punct(
+                candidate = _extract_with_trailing_punct(
                     doc_index.doc_text,
                     int(ctx_start),
                     int(ctx_end),
                 )
+                if _span_matches_hash(row, candidate, "context"):
+                    context = candidate
 
         if not definition and pdf_path is not None and pdf_token_index:
             spec = _select_hash_specs(row, "definition")

scripts/prepare_defextra_legal.py

@@ -524,7 +524,12 @@ def main() -> None:
                             for spec in (def_tail_spec, def_tail_alt_spec)
                             if spec
                         ]
-                        if token_index and expected_len and head_specs and tail_specs:
+                        if (
+                            token_index
+                            and expected_len
+                            and head_specs
+                            and tail_specs
+                        ):
                             for head in head_specs:
                                 for tail in tail_specs:
                                     anchor_span = _find_span_by_anchors(
@@ -613,7 +618,12 @@ def main() -> None:
                             for spec in (ctx_tail_spec, ctx_tail_alt_spec)
                             if spec
                         ]
-                        if token_index and expected_len and head_specs and tail_specs:
+                        if (
+                            token_index
+                            and expected_len
+                            and head_specs
+                            and tail_specs
+                        ):
                             for head in head_specs:
                                 for tail in tail_specs:
                                     anchor_span = _find_span_by_anchors(

scripts/report_defextra_status.py

@@ -76,9 +76,15 @@ def _index_recent_pdfs(
             keys = {stem, stem.lower(), normalize_paper_id(stem)}
             if stem.startswith("paper_"):
                 stripped = stem[len("paper_") :]
-                keys.update({stripped, stripped.lower(), normalize_paper_id(stripped)})
+                keys.update(
+                    {stripped, stripped.lower(), normalize_paper_id(stripped)},
+                )
             if stem.endswith("_fixed") or stem.endswith("-fixed"):
-                base = stem[: -len("_fixed")] if stem.endswith("_fixed") else stem[: -len("-fixed")]
+                base = (
+                    stem[: -len("_fixed")]
+                    if stem.endswith("_fixed")
+                    else stem[: -len("-fixed")]
+                )
                 if base:
                     keys.update({base, base.lower(), normalize_paper_id(base)})
             match = arxiv_re.match(stem)
@@ -112,13 +118,17 @@ def main() -> None:
     parser.add_argument(
         "--legal-report",
         type=Path,
-        default=Path("results/paper_results/defextra_legal_tablefix_report.txt"),
+        default=Path(
+            "results/paper_results/defextra_legal_tablefix_report.txt",
+        ),
         help="Report generated by prepare_defextra_legal.py.",
     )
     parser.add_argument(
         "--hydrated-csv",
         type=Path,
-        default=Path("results/paper_results/defextra_hydrated_tablefix_test.csv"),
+        default=Path(
+            "results/paper_results/defextra_hydrated_tablefix_test.csv",
+        ),
         help="Hydrated CSV from hydrate_defextra.py.",
     )
     parser.add_argument(
@@ -142,10 +152,16 @@ def main() -> None:
     args = parser.parse_args()
 
     legal_rows = _load_csv(args.legal_csv)
-    hydrated_rows = _load_csv(args.hydrated_csv) if args.hydrated_csv.exists() else []
+    hydrated_rows = (
+        _load_csv(args.hydrated_csv) if args.hydrated_csv.exists() else []
+    )
 
-    ref_ids = {row.get("paper_id", "") for row in legal_rows if row.get("paper_id")}
-    hyd_ids = {row.get("paper_id", "") for row in hydrated_rows if row.get("paper_id")}
+    ref_ids = {
+        row.get("paper_id", "") for row in legal_rows if row.get("paper_id")
+    }
+    hyd_ids = {
+        row.get("paper_id", "") for row in hydrated_rows if row.get("paper_id")
+    }
     missing_papers = sorted(ref_ids - hyd_ids)
 
     missing_defs, missing_ctxs = _parse_missing_report(args.legal_report)
@@ -163,7 +179,11 @@ def main() -> None:
         except ValueError:
             continue
         row = idx.get((pid, concept))
-        if row and (row.get("definition_type") or "").strip().lower() == "implicit":
+        if (
+            row
+            and (row.get("definition_type") or "").strip().lower()
+            == "implicit"
+        ):
             implicit_defs.append(item)
     for item in missing_ctxs:
         try:
@@ -171,7 +191,11 @@ def main() -> None:
         except ValueError:
             continue
         row = idx.get((pid, concept))
-        if row and (row.get("definition_type") or "").strip().lower() == "implicit":
+        if (
+            row
+            and (row.get("definition_type") or "").strip().lower()
+            == "implicit"
+        ):
             implicit_ctxs.append(item)
 
     cutoff_ts = time.time() - (args.recent_days * 86400)
@@ -206,31 +230,35 @@ def main() -> None:
     for pid in missing_papers:
         lines.append(f"- {pid}")
     lines.append("")
-    lines.append(f"Missing definition spans marked implicit: {len(implicit_defs)}")
+    lines.append(
+        f"Missing definition spans marked implicit: {len(implicit_defs)}",
+    )
     for item in implicit_defs:
         lines.append(f"- {item}")
     lines.append("")
-    lines.append(f"Missing context spans marked implicit: {len(implicit_ctxs)}")
+    lines.append(
+        f"Missing context spans marked implicit: {len(implicit_ctxs)}",
+    )
     for item in implicit_ctxs:
         lines.append(f"- {item}")
     lines.append("")
     lines.append(
         f"Missing papers with recent PDFs (<= {args.recent_days} days): "
-        f"{len(recent_missing_papers)}"
+        f"{len(recent_missing_papers)}",
     )
     for pid in recent_missing_papers:
         lines.append(f"- {pid}")
     lines.append("")
     lines.append(
         f"Missing definition spans with recent PDFs (<= {args.recent_days} days): "
-        f"{len(recent_missing_defs)}"
+        f"{len(recent_missing_defs)}",
     )
     for item in recent_missing_defs:
         lines.append(f"- {item}")
     lines.append("")
     lines.append(
         f"Missing context spans with recent PDFs (<= {args.recent_days} days): "
-        f"{len(recent_missing_ctxs)}"
+        f"{len(recent_missing_ctxs)}",
     )
     for item in recent_missing_ctxs:
         lines.append(f"- {item}")