Oct 2026· Proceedings of the Workshop on the ACM RecSys Challenge· 1 citation· ⚡ 1 influential· 1 references
Abstract
We present suryaaseran1, our conversational music recommender for the TalkPlayData Challenge (ACM RecSys 2026) that treats each dialogue turn’s user reactions as a preference-elicitation signal for both retrieval and ranking. The dataset’s goal-progress feedback is delayed by one turn and includes rejected tracks logged as if they were positives; we align the signal in time, remove rejected-track contamination from training, and reconstruct it from raw dialogue when it is withheld at test time. This signal drives a role-tagged dialogue anchor, encoding prior turns, recommendations, and per-turn reactions, behind a LoRA fine-tune of a 279M-parameter multilingual encoder, the system’s only trained neural component. Retrieval around this anchor is a union of nine inexpensive sources arbitrated by a gradient-boosted learning-to-rank model over 67 features, with multi-source agreement the strongest single feature. The full pipeline trains and serves on a single consumer desktop with no discrete GPU, and reaches a composite score of 0.49 on the blind generalization set.
Team npatta01’s submission to the RecSys Challenge 2026 conversational music recommendation task is described and failure cases from the submitted run show extracted constraints the pipeline could not enforce.
Nidhin Pattaniyil, Semih Yagli, Tanwir Zaman· Proceedings of the Workshop...· 1 citation· ⚡1
This paper describes the solution submitted by team Hallucinated to the ACM RecSys Challenge 2026, based on the TalkPlayData conversational music recommendation dataset. The Challenge poses a dual task: at every turn of a multi-turn dialogue, a system must both (i) rank the most relevant music tracks from the catalogue...
Abdallah Alkhetiar, Luigi Inguaggiato, Nicolò Locatelli et al.· Proceedings of the Workshop...· 1 citation
The ACM-RecSys 2026 Music Conversational Recommendation Challenge (Music-CRS) bridges natural language processing and recommender systems, requiring systems to engage in multi-turn dialogues and recommend relevant tracks from a large catalog. In this paper, we present Team Sargam’s solution, a robust hybrid retrieval a...
Aditya Rai, Venkatesh Shukla, Sanskar Aggarwal et al.· Proceedings of the Workshop...· 1 citation· ⚡1
The RecSys Challenge 2026 studies conversational music recommendation as a joint item recommendation and response generation problem: given a multi-turn dialogue, systems must retrieve relevant tracks from a large catalog and produce a grounded natural-language response. This paper presents the challenge task, dataset,...
Seungheon Doh, Sergio Oramas, B. Sguerra et al.· Proceedings of the Workshop...· 0 citations
The RecSys Challenge 2026 Music-CRS (TalkPlay) task formalizes this as two coupled sub-problems: given dialogue history and user context, retrieve a ranked list of the top-20 tracks from the full, unrestricted catalog, and generate a response that justifies the recommendation while sustaining conversational coherence.
Simran Sundrani, Mohan Bhambhani· Proceedings of the Workshop...· 1 citation· ⚡1
This diagnostic compares the full pipeline with reciprocal-rank fusion, ablate per-retriever features, and decompose ranking error into retrieval misses, reranking exclusions, and ranks-2–20 ordering loss and complement the leaderboard result with a diagnostic that fits on Train and evaluates the official Devset.
Ryohei Wakatsuki· Proceedings of the Workshop...· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
