arXiv:2507.03047v1 Announce Type: cross Abstract: Recent advances have applied large language models (LLMs) to sequential recommendation, leveraging their pre-training knowledge and reasoning capabilities to provide more personalized user experiences. However, existing LLM-based methods fail to sufficiently leverage the rich temporal information inherent in users' historical interaction sequences, stemming from fundamental architectural constraints: LLMs process information through self-attention mechanisms that lack inherent sequence ordering and rely on position embeddings designed primarily for natural language rather than user interaction sequences. This limitation significantly impairs their ability to capture the evolution of user preferences over time and predict future interests accurately. To address this critical gap, we propose Counterfactual Enhanced Temporal Framework for LLM-Based Recommendation (CETRec). CETRec is grounded in causal inference principles, which allow it to isolate and measure the specific impact of temporal information on recommendation outcomes. By conceptualizing temporal order as an independent causal factor distinct from item content, we can quantify its unique contribution through counterfactual reasoning--comparing what recommendations would be made with and without temporal information while keeping all other factors constant. This causal framing enables CETRec to design a novel counterfactual tuning objective that directly optimizes the model's temporal sensitivity, teaching LLMs to recognize both absolute timestamps and relative ordering patterns in user histories. Combined with our counterfactual tuning task derived from causal analysis, CETRec effectively enhances LLMs' awareness of both absolute order (how recently items were interacted with) and relative order (the sequential relationships between items).
