Scholarly Writing Refiner

# Scholarly Writing Refiner

Polish academic English paragraph by paragraph — grammar, word choice, voice, coherence, and sentence structure — according to international academic journal standards. Output revision suggestions alongside the polished text. Use for manuscript editing, journal submission editing, thesis/dissertation polishing, and "make this paragraph sound more academic" requests.

## Quick Start

Provide: the text to polish, and optionally the paper type (journal article / conference paper / thesis / review), the target journal or field (Nature, IEEE, AAAI, a specific field), and the polishing focus (comprehensive / grammar only / word choice only / coherence only).

## Review Dimensions

Polishing runs across five independently-rated dimensions: **Grammar** (subject-verb agreement, tense, articles, prepositions, clause structure, punctuation), **Word Choice** (academic register, precision, collocation, avoiding colloquialisms), **Voice** (active/passive selection, tense consistency), **Coherence & Cohesion** (transitions within and between paragraphs, the argumentation chain, signal words), and **Sentence Structure** (variety, balance of long and short sentences, coordination and subordination).

## Grammar Review

### Common Errors Checklist

| Error Type | Incorrect | Correction | Why |
|---|---|---|---|
| Subject-verb disagreement | "The results of the experiment **shows**..." | "...**show**..." | Subject is "results" (plural) |
| Missing/misused article | "We propose **method** to solve..." | "We propose **a method** to solve..." | Singular countable nouns need an article |
| Dangling modifier | "**Using the proposed method,** the accuracy was improved." | "**Using the proposed method, we** improved the accuracy." | The implied subject must match the main clause's subject |
| Run-on sentence | "The model performs well**,** it achieves 95% accuracy." | "...well**;** it achieves..." or split into two sentences | A comma can't join two independent clauses |
| Incomplete comparison | "Our method is **more efficient**." | "...more efficient **than the baseline**." | Comparatives need an explicit object |
| Broken parallelism | "The system can **detect, classify,** and **is able to segment**..." | "...**detect, classify, and segment**..." | Coordinated elements need the same grammatical form |
| that/which confusion | "The model **which** we proposed..." | "The model **that** we proposed..." | Restrictive relative clauses use "that" |
| Irregular plurals | "These **phenomenon** indicate..." | "These **phenomena** indicate..." | Watch for irregular plural forms |

### Punctuation

- **Serial (Oxford) comma**: "A, B, and C" — recommended in academic writing.
- **Em dash**: consistent spacing per the target journal's style.
- **Capitalization after a colon**: capitalize only if a complete sentence follows.
- **Quotation marks and periods**: American style puts the period inside quotes; British style outside — match the target journal's regional convention.
- **Abbreviation periods**: "e.g.," / "i.e.," / "et al." — note the required comma.

### Tense by Section

| Section | Tense | Example |
|---|---|---|
| Abstract | Past (what was done) + present (conclusions) | "We **proposed** a method... The results **show** that..." |
| Introduction | Present (current consensus) + past (prior work) | "Deep learning **has become**... Smith et al. **demonstrated** that..." |
| Methods | Past | "We **trained** the model on... The data **were** preprocessed..." |
| Results | Past | "The model **achieved** 95% accuracy. Table 2 **shows**..." |
| Discussion | Present (interpretation) + past (citing results) | "This result **suggests** that... Our findings **indicated** that..." |
| Conclusion | Past (summary) + present (contribution/significance) | "We **proposed** and **evaluated**... This work **contributes** to..." |

## Word Choice

**Colloquial → academic**: "a lot of" → numerous/a substantial number of; "get" → obtain/acquire/achieve; "show" → demonstrate/illustrate/indicate/reveal (pick by nuance — demonstrate emphasizes proof, indicate emphasizes suggestion); "big/huge" → substantial/significant; "use" → employ/utilize/leverage/adopt; "try" → attempt; "look at" → examine/investigate/analyze; "find out" → determine/identify; "point out" → highlight/emphasize; "deal with" → address/tackle/mitigate; "make sure" → ensure/verify; "need" → require/necessitate. Don't over-substitute — "can" is acceptable in academic writing as-is.

**Vague → precise**: "very good results" → the actual statistic ("a 12% increase in accuracy"); "some researchers" → the actual citations; "recently" → an actual time range ("since 2020"); "a few" → the actual count ("n=3"); "it is known that" → "Prior work has established that (citation)"; "this is important" → state *why* it matters.

**Reducing redundancy**: "in order to" → "to"; "due to the fact that" → "because"; "it is worth noting that" → "Notably," or delete; "the vast majority of" → "most"; "has the ability to" → "can"; "take into consideration" → "consider"; "is in agreement with" → "agrees with."

## Voice

| Scenario | Recommended Voice | Example |
|---|---|---|
| Describing the authors' actions | Active ("We") | "**We** trained the model using..." |
| Describing general methods/established facts | Passive | "The data **were** collected from..." |
| Emphasizing the object of an action | Passive | "The samples **were analyzed** using mass spectrometry." |
| Reporting results | Prefer active | "**Our method achieves** 95% accuracy." |
| Describing equipment/materials | Passive | "The solution **was heated** to 100°C." |

**Common voice issues**: overuse of passive ("It was found by us that the results were improved by the method" → "We found that our method improved the results"); inconsistent person across the paper (pick "We" or "The authors" and stay consistent); meaningless passive ("It can be seen that accuracy increases" → "Accuracy increases").

**Person conventions**: "We" is the most common choice, accepted by many journals even for single-author papers; "The authors" is a more formal alternative some journals prefer; "I" is used in single-author theses/dissertations (not accepted everywhere); "One" is generic/hypothetical and reads as somewhat old-fashioned.

## Coherence and Cohesion

**Intra-paragraph signal words by relationship**: Addition (furthermore, moreover, additionally); Contrast (however, in contrast, conversely, nevertheless); Cause & Effect (therefore, consequently, as a result, hence); Exemplification (for example, specifically, in particular); Emphasis (indeed, notably, importantly); Concession (although, despite, while); Summary (in summary, overall); Sequence (first, subsequently, then); Condition (if, provided that, given that).

**Inter-paragraph transition patterns**: Hook (the end of one paragraph raises the question the next answers); Recap ("Having established X, we now turn to..."); Contrast Bridge ("While these methods achieve reasonable accuracy, they fail to address..."); Question Bridge ("How can we overcome this limitation? In this section..."); Topic Sentence (the first sentence of each paragraph states its core argument).

**Common coherence problems**: jumping argumentation (leaping from A to C without the B step — add the missing reasoning); signal-word overuse (starting every sentence with "However"/"Moreover" — reduce and let sentence structure carry the logic instead); signal-word misuse (using "Furthermore" to express contrast instead of "However"); overly long paragraphs (past 8–10 sentences — split by argument point); overly short paragraphs (1–2 sentences — merge or expand); unclear reference ("This shows..." — say what "this" is: "This result shows...").

## Sentence Structure

**Variety strategies**: participial-phrase opening ("We use attention, and we improve accuracy" → "Leveraging attention, we improve accuracy"); inversion for emphasis; appositive insertion ("The model, which was proposed by Smith, achieves..." → "The model, proposed by Smith (2023), achieves..."); nominalization for a completed action as a subject; parallel structure for a list of properties ("The method is fast. It is also accurate. It is scalable too." → "The method is fast, accurate, and scalable"); fronted adverbial.

**Problems to avoid**: sentences over ~40 words (split into 2–3); several consecutive short, choppy sentences (combine into one with the right subordination); starting with "There is/are" ("There are many studies that focus on..." → "Many studies focus on..."); overusing "It is...that" cleft sentences; noun pile-ups ("deep learning image classification model performance" → "the performance of a deep learning model for image classification").

## Section-Specific Polishing

**Abstract** — typically 150–300 words (follow the target journal); structure: background (1–2 sentences) → problem/motivation (1 sentence) → method (2–3 sentences) → results (1–2 sentences) → significance (1 sentence). Past tense for describing the work, present tense for conclusions. No citations, no undefined abbreviations, no figure/table numbers.

**Introduction** — the classic funnel: broad context → specific problem → existing methods and their limitations → proposed method/contributions → paper outline. Every paragraph needs a clear topic sentence; stay objective reviewing prior work, never disparaging. Common openers: "In recent years, ... has attracted increasing attention." / "Despite significant progress, ... remains a challenge." / "To address this issue, we propose..."