Multi-Modal Semantic Expansion with Constrained LLM Reranking for Conversational Music Recommendation

• URL: http://arxiv.org/abs/2608.23484v1
• Date: Mon, 24 Aug 2026 16:48:50 GMT
• Title: Multi-Modal Semantic Expansion with Constrained LLM Reranking for Conversational Music Recommendation
• Abstract summary: We present Team Semiintelligencn's solution for the ACM RecSys 2026 TalkPlayData Challenge.<n>Our submitted system employs a three-stage pipeline: (1) multi-modal retrieval constructing decay-weighted centroids across seven dense embedding spaces.<n>We optimize RRF weights on a 500-session development split via differential evolution, improving MRR by +19.5%.
• Score: 2.8110777003480654
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We present Team Semiintelligencn's solution for the ACM RecSys 2026 TalkPlayData Challenge, addressing conversational music recommendation through a multi-modal and personalized conversational recommender system. Our submitted system employs a three-stage pipeline: (1) multi-modal retrieval constructing decay-weighted centroids across seven dense embedding spaces - track- and user-level CF-BPR, Qwen3 (metadata, lyrics, attributes), CLAP audio, and SigLIP visual - supplemented by BM25 lexical retrieval and an artist substring-match signal, all fused via weighted Reciprocal Rank Fusion (RRF) with optimized signal weights; (2) lightweight reranking (history filtering, popularity smoothing, and catalog diversity penalization); and (3) persona-diversified response generation using GPT-4o-mini. Beyond this submitted configuration, we report development-time experiments with additional components - constrained LLM-guided artist injection, album continuation signals, XGBoost LambdaMART, and a superior GPT-4.1 response prompt - that were not deployed to Blind B due to cost and complexity constraints. We optimize RRF weights on a 500-session development split via differential evolution, improving MRR by +19.5%. On Blind A, we observe that unconstrained LLM-guided injection across 54 sessions causes catastrophic nDCG regression (-18.9%), while conservative injection on only 9 sessions yields the best observed Blind A nDCG - a finding we present as a Blind A observation warranting further validation. The submitted system achieves a Blind B composite score of 0.3213.

Related papers

• Retrieval-Augmented Multi-Prompt Ensemble for Minor-Grain Breeding Information Extraction [45.40972888157319]
  RAME (Retrieval-Augmented Multi-Prompt Ensemble) is a training-free framework that elicits multiple LLM outputs under controlled diversity.<n> RAME achieves a Total Score of 0.499 (NER 0.730, RE 0.346) on the leaderboard, ranking 1st and surpassing the official Track-A baseline powered by GPT-5.5 (0.448)
  arXiv  Detail & Related papers  (2026-09-07T07:30:49Z)
• Sign Language Video Synthesis via Loss-Guided Multi-Expert GANs [5.455881292140741]
  This report presents a framework for sign language video synthesis using a loss-guided multi-expert Generative Adversarial Network (GAN)<n>Three specialized discriminators -- global, hand, and head -- each guide a corresponding expert branch in the generator toward a distinct visual region.<n>The system was showcased at the 2025 Hong Kong Frontier Technology Summit.
  arXiv  Detail & Related papers  (2026-08-13T15:32:40Z)
• Multimodal Music Recommendation System using LLMs [53.1178230380919]
  We propose a framework for session-based music recommendation that enriches the LastFM-1K dataset with three complementary signals.<n>We adopt the E4SRec framework by extending it with multimodal features and different item ID encoder backbones.<n>Our experiments show that integrating content-based features improves over ID-only baselines up to 95% in terms of Recall and 79% in terms of NDCG.
  arXiv  Detail & Related papers  (2026-05-28T15:34:12Z)
• MARQUIS: A Three-Stage Pipeline for Video Retrieval-Augmented Generation [64.33427079838602]
  Retrieval-augmented generation from videos requires systems to retrieve relevant audiovisual evidence from large corpora and synthesize it into coherent, attributed text.<n>We present MARQUIS: a three-stage pipeline that addresses these limitations through query expansion, fusion, and reranking.<n>For article generation, ITER-QA-BASE improves average human score from 3.09 to 3.83 over the CAG baseline, while MARQUIS-RLM achieves a human score of 3.30 and the strongest citation recall among non-QA systems.
  arXiv  Detail & Related papers  (2026-05-17T20:19:04Z)
• RLearner-LLM: Balancing Logical Grounding and Fluency in Large Language Models via Hybrid Direct Preference Optimization [5.0009109610311855]
  We propose RLearner-LLM with Hybrid-DPO.<n>RLearner-LLM yields up to 6x NLI improvement over SFT.<n>On Gemma 4 E4B-it, Hybrid-DPO lifts NLI in four of five domains.
  arXiv  Detail & Related papers  (2026-05-06T06:36:09Z)
• RefineBench: Evaluating Refinement Capability of Language Models via Checklists [71.02281792867531]
  We evaluate two refinement modes: guided refinement and self-refinement.<n>In guided refinement, both proprietary LMs and large open-weight LMs can leverage targeted feedback to refine responses to near-perfect levels within five turns.<n>These findings suggest that frontier LMs require breakthroughs to self-refine their incorrect responses.
  arXiv  Detail & Related papers  (2025-11-27T07:20:52Z)
• LumiCRS: Asymmetric Contrastive Prototype Learning for Long-Tail Conversational Recommender Systems [2.4786559575079083]
  Only 10% of head movies account for nearly half of all mentions, whereas about 70% of tail movies receive merely 26% of the attention.<n>This imbalance gives rise to three critical challenges: head over-fitting, body representation drift, and tail sparsity.<n>We propose LumiCRS, an end-to-end framework that adjusts long-tail imbalance through three mutually reinforcing layers.
  arXiv  Detail & Related papers  (2025-07-07T07:33:00Z)
• Reverse Preference Optimization for Complex Instruction Following [61.39734201711077]
  We propose a simple yet effective method called Reverse Preference Optimization (RPO)<n>It mitigates noise in preference pairs by dynamically reversing the constraints within the instruction to ensure the chosen response is perfect.<n>RPO scales effectively across model sizes, with the 70B RPO model surpassing GPT-4o.
  arXiv  Detail & Related papers  (2025-05-28T09:44:27Z)
• MLCA-AVSR: Multi-Layer Cross Attention Fusion based Audio-Visual Speech Recognition [62.89464258519723]
  We propose a multi-layer cross-attention fusion based AVSR approach that promotes representation of each modality by fusing them at different levels of audio/visual encoders. Our proposed approach surpasses the first-place system, establishing a new SOTA cpCER of 29.13% on this dataset.
  arXiv  Detail & Related papers  (2024-01-07T08:59:32Z)
• Low-complexity deep learning frameworks for acoustic scene classification [64.22762153453175]
  We present low-complexity deep learning frameworks for acoustic scene classification (ASC) The proposed frameworks can be separated into four main steps: Front-end spectrogram extraction, online data augmentation, back-end classification, and late fusion of predicted probabilities. Our experiments conducted on DCASE 2022 Task 1 Development dataset have fullfiled the requirement of low-complexity and achieved the best classification accuracy of 60.1%.
  arXiv  Detail & Related papers  (2022-06-13T11:41:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.