EMNLP workshop any good? Or any other NLP venue good for VLM eval work? [D]

A PhD student's candid post about a paper rejection has gone viral on an AI research forum, highlighting a common but painful crossroads in academic publishing. The student's work on Vision-Language Model (VLM) evaluation was rejected from a top-tier (A*) imaging conference, with reviewers citing a lack of clinical validation and deeming it better suited for an NLP venue. This has left the researcher, who is openly targeting industry roles post-PhD and desires prestigious 'A*' publications, weighing a critical strategic decision: submit to a highly relevant workshop at the EMNLP conference or aim for a lower-tier imaging venue.

The post has ignited a substantial discussion among researchers about the perceived value and impact of workshops at major conferences like EMNLP, ACL, or NeurIPS. Many senior commenters are advising that a well-regarded workshop can be an excellent venue for focused feedback and community building, especially for niche interdisciplinary work like VLM eval that falls between computer vision and NLP. The consensus advice leans towards targeting the specialized workshop for faster publication and networking, rather than chasing a potentially ill-fitting journal or conference tier, as industry recruiters often value the technical substance and relevance over the venue's ranking alone.

This scenario underscores a broader tension in AI research as fields like multimodal AI blur traditional disciplinary lines. Researchers developing tools that straddle vision and language—such as models that evaluate GPT-4V or LLaVA—increasingly face challenges in identifying the right peer community. The viral thread serves as a real-time case study in academic career strategy, offering a crowdsourced guide on navigating publication venues, the trade-offs between prestige and fit, and how to build a profile for a successful transition into industry AI roles.