Papers for

dialogue system developers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Locus cuts language model output length without hurting preference quality

LOCUS: Task-Aware Low-Rank Post-Training for Token-Efficient Language Generation

Abstract: Large language model serving costs scale directly with output sequence length, yet standard preference alignment often inflates response verbosity without improving utility. We study whether the parameterization of post-training updates affects generation length: low-rank subspaces alter sequence length without modifying the alignment loss. We present LOCUS, a method that selects a task-aware low-rank adaptation subspace to minimize output-token cost subject to a utility constraint. Within this subspace, post-training retains the native preference objective with a frozen backbone. Across Anthropic HH-RLHF dialogue preferences, we evaluate two $\sim$3B decoder backbones, Pythia-2.8B and Qwen2.5-3B, against protocol-matched full-parameter DPO and DrDPO branches and the released SamPO checkpoint. LOCUS reduces continuation length by up to 39.84\% on Pythia-2.8B and by 14.87--17.58\% on Qwen2.5-3B while updating only 0.24--0.28\% of model parameters, with no material change in the internal preference diagnostic.

Thu 10 SeptComputation and LanguageArtificial IntelligenceMachine Learning
The gist
Large language models often produce longer answers than needed, which increases costs and time to get results. The authors show that by carefully adjusting just a small part of the model using a special low-rank method, they can make the model give shorter replies while keeping the answers just as good. This technique works by changing how the model generates words without changing its original goals. Their method, called LOCUS, was tested on two medium-sized language models and successfully reduced the length of responses by up to 40% without lowering quality.
Open → 2609.11739v1

Multimodal model tracks group emotions every second in dialogues

Multimodal Temporal Modeling for Continuous Group Emotion Recognition in Multi-party Dialogues

Abstract: To realize natural behavior in dialogue agents in multi-party dialogue scenarios, it is important to understand group emotion such as valence and arousal as a whole. Most prior work addressed this task at the utterance level or using a coarse-grained time window, which is not sufficient to capture emotional dynamics. In this study, we formulate continuous recognition of the Group Emotion at a one-second resolution. Moreover, we also introduce the Mixed state, which captures the emotional divergence among participants in the group. We constructed a dataset with frame-level soft labels based on the TEIDAN corpus and propose a multimodal temporal framework that integrates audio and video information using a sliding-window context. Experimental results demonstrate that the temporal Transformer outperforms simple baselines and shows stronger temporal agreement with the ground-truth labels than the LLM-based model. The effect of context length is limited, whereas audio-visual input outperforms either unimodal input on the continuous-label metrics. Additionally, our analysis shows larger Group Emotion recognition errors in intervals with high Mixed values, exposing emotional divergence as a key challenge for group emotion recognition.

Thu 10 SeptMultimedia
The gist
Understanding how a group feels during a conversation can help make dialogue systems more natural. The authors created a method to recognize group emotions like mood and energy every second using both sound and video. They also introduced a new idea called the Mixed state, which shows when group members feel differently. Their approach works better than some existing models but finds it harder to detect emotions when the group’s feelings diverge a lot.
Open → 2609.11164v1