Papers for

email client developers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Personal email agents compared reveal limits of rules and few-shot llm prompts

Measure Before You Train: Typed Decision Models vs. a Classical Baseline vs. a Generative LLM for Routing a Personal Multilingual Inbox

Abstract: Personal email agents decide, for every message, what it is, whether it is urgent, whether to draft a reply, create a task, or ignore it. Their error rate is rarely measured: the stored labels were written by the system itself. We present Foyer, a self-hosted personal inbox agent in daily use, and a measurement protocol (stratified sampling, frozen snapshots, thread-grouped splits, pre-registered hypotheses, human labels in the agent's review loop) that yields ground truth before any training. On 739 real Italian/English emails labelled by the owner (231 unbiased, 500 adversarially sampled), we compare nine systems under thread-grouped cross-validation: the production keyword rules, a TF-IDF + logistic-regression baseline, local typed decision models (Tev1, Nimble), a hosted decision model (Nimble v4-max), and hosted generative LLMs (Qwen2.5-7B, Gemma 4 26B/31B). Four results. The production rules score 28% on message type against a 58% majority class: the field shown most prominently carried no information. With ~200 labels, logistic regression has the highest accuracy on every field (type 0.79, urgency 0.74, needs-reply 0.83, is-noise 0.90), though it never predicts the two rarest classes and is statistically tied with Gemma 4 on type; few-shot prompting does not close the gap. The sender's relationship (client, supplier, service) is recovered from other emails of the same contact with 98.6% accuracy where the contact has been seen, and is not inferable from a single message by any model: it belongs in the contact graph, not the prompt. Finally, accuracy is not the product metric: composed into actions, the most accurate system most often ignores emails needing action (18% vs 4% for the best LLM); a cost-chosen threshold on its noise probability halves that rate and makes it the cheapest system.

Fri 23 OctComputation and LanguageHuman-Computer InteractionMachine Learning
The gist
Deciding how to handle every email, like whether it needs a reply or can be ignored, is a hard task for personal email assistants. The authors built a system called Foyer to test and compare different ways to sort emails—including simple rules, traditional machine learning, typed decision models, and large language models. They found that simple keyword rules performed poorly and that a basic machine learning model often worked best with enough training data. They also discovered some information about the sender's relationship is hard to guess from just one message and is better tracked as part of the contacts' history. Finally, effectiveness isn’t just about accuracy—how the system acts on emails matters too, with trade-offs on missing important actions versus wrongly flagging noise.
Open → 2610.32955v1

AI shifts cultural styles in professional email writing

AI-Generated Email Drafts Shift Culturally Distinctive Communication Styles in Professional Email

Abstract: AI assistants that support email composition may shift cultural communication norms, such as the directness typical of low-context cultures like the US versus the indirectness and contextual sensitivity central to high-context cultures like Japan. Yet it remains unknown to what extent people adopt and edit AI drafts inconsistent with their cultural communication norms. We address this through a preregistered within-subject experiment in which Japanese and American participants wrote workplace emails in their native language without AI, with a low-context AI, and with a high-context AI. We found that Japanese participants wrote emails with significantly more high-context markers (politeness, apologies) than Americans. But AI drafts shifted participants' emails toward the draft's style, with larger shifts when the draft was culturally misaligned: Japanese drifted most under low-context drafts, Americans most under high-context drafts. These findings suggest AI drafts risk overwriting cultural communication norms unless they adapt to users' communication styles.

Tue 22 SeptHuman-Computer Interaction
The gist
People from different cultures write emails in different ways—some are direct, like in the US, while others are more polite and indirect, like in Japan. This paper shows that when people use AI to help write emails, the AI’s style can shift how they write. For example, Japanese users became more direct when using an AI trained on American styles, and Americans became more polite with a Japanese-style AI. The authors suggest that AI tools should adapt better to the user's cultural style to avoid changing how people communicate.
Open → 2609.26403v1