Four CHI '26 papers I wish I wrote

A researcher describes four CHI ’26 papers that sparked strong “I wish I wrote this” reactions and unpacks what made each compelling. The first paper reports a large between-subjects experiment where participants used an LLM-based autocomplete tool to write evaluations of male and female résumé presenters; counter-stereotypical suggestions raised perceived competence and salary offers for the female candidate but triggered backlash in likability and did not overturn overall male preference. The researcher connects this to prior work on reducing gender bias in recommendation letters, praises the integration of debiasing into task performance, flags ethical trade-offs, and notes practical limits (lack of experimental collaborators) while proposing follow-ups like gender-blind résumé manipulation and LLM-generated suggestion tools.

The second paper analyzes 405 Reddit help-seeking posts about scams, identifying five emotional drivers (fear, hope, trust, guilt, belonging), four help needs (sensemaking, guidance, emotional support, action), and risk factors such as financial precarity and mental-health vulnerabilities, recommending just-in-time interventions. The third paper presents a values-linked time-tracking prototype tested with 15 participants; annotating activities against personal values prompted reflection and some behavior change and motivates episodic “life studies” rather than continuous self-tracking. Across these pieces the researcher sees opportunities to translate persuasive-technology methods into ethical interventions for hiring, scam prevention, and meaningful self-tracking.