# Self-Supervised and Controlled Multi-Document Opinion Summarization

arXiv: https://arxiv.org/abs/2004.14754 · HTML: https://mlsanity.com/paper/2004.14754/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Apr 2020.

## Built on by (most cited first, 23 of 23)

- [Aspect-Controllable Opinion Summarization](https://mlsanity.com/paper/2109.03171/) (2021)
- [AI Hallucinations: A Misnomer Worth Clarifying](https://mlsanity.com/paper/2401.06796/) (2024)
- [Overview of the 2021 Key Point Analysis Shared Task](https://mlsanity.com/paper/2110.10577/) (2021)
- [Convex Aggregation for Opinion Summarization](https://mlsanity.com/paper/2104.01371/) (2021)
- [Self-Supervised Multimodal Opinion Summarization](https://mlsanity.com/paper/2105.13135/) (2021)
- [Comparative Opinion Summarization via Collaborative Decoding](https://mlsanity.com/paper/2110.07520/) (2021)
- [Efficient Few-Shot Fine-Tuning for Opinion Summarization](https://mlsanity.com/paper/2205.02170/) (2022)
- [One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation](https://mlsanity.com/paper/2402.11683/) (2024)
- [Scientific Opinion Summarization: Paper Meta-review Generation Dataset, Methods, and Evaluation](https://mlsanity.com/paper/2305.14647/) (2023)
- [Beyond Opinion Mining: Summarizing Opinions of Customer Reviews](https://mlsanity.com/paper/2206.01543/) (2022)
- [Product Description and QA Assisted Self-Supervised Opinion Summarization](https://mlsanity.com/paper/2404.05243/) (2024)
- [OpinioRAG: Towards Generating User-Centric Opinion Highlights from Large-scale Online Reviews](https://mlsanity.com/paper/2509.00285/) (2025)
- [Noisy Pairing and Partial Supervision for Stylized Opinion Summarization](https://mlsanity.com/paper/2211.08723/) (2022)
- [OpineSum: Entailment-based self-training for abstractive opinion summarization](https://mlsanity.com/paper/2212.10791/) (2022)
- [Summarization of Opinionated Political Documents with Varied Perspectives](https://mlsanity.com/paper/2411.04093/) (2024)
- [Unsupervised Opinion Summarisation in the Wasserstein Space](https://mlsanity.com/paper/2211.14923/) (2022)
- [Simple Yet Effective Synthetic Dataset Construction for Unsupervised Opinion Summarization](https://mlsanity.com/paper/2303.11660/) (2023)
- [Fine-Grained Opinion Summarization with Minimal Supervision](https://mlsanity.com/paper/2110.08845/) (2021)
- [Distilling Opinions at Scale: Incremental Opinion Summarization using XL-OPSUMM](https://mlsanity.com/paper/2406.10886/) (2024)
- [LFOSum: Summarizing Long-form Opinions with Large Language Models](https://mlsanity.com/paper/2410.13037/) (2024)
- [AaKOS: Aspect-adaptive Knowledge-based Opinion Summarization](https://mlsanity.com/paper/2306.05537/) (2023)
- [Query-Guided Self-Supervised Summarization of Nursing Notes](https://mlsanity.com/paper/2407.04125/) (2024)
- [LLMs as Architects and Critics for Multi-Source Opinion Summarization](https://mlsanity.com/paper/2507.04751/) (2025)

## Abstract (arXiv)

We address the problem of unsupervised abstractive summarization of collections of user generated reviews with self-supervision and control. We propose a self-supervised setup that considers an individual document as a target summary for a set of similar documents. This setting makes training simpler than previous approaches by relying only on standard log-likelihood loss. We address the problem of hallucinations through the use of control codes, to steer the generation towards more coherent and relevant summaries.Finally, we extend the Transformer architecture to allow for multiple reviews as input. Our benchmarks on two datasets against graph-based and recent neural abstractive unsupervised models show that our proposed method generates summaries with a superior quality and relevance.This is confirmed in our human evaluation which focuses explicitly on the faithfulness of generated summaries We also provide an ablation study, which shows the importance of the control setup in controlling hallucinations and achieve high sentiment and topic alignment of the summaries with the input reviews.