SeDT: Sentence-Transformer Decision-Transformer Conditioning for Multi-Turn Conversation Reliability
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910260374011904 |
|---|---|
| author | Setti, Ramakrishna Vamsi Rachapudi, Jagadeesh Chaudhary, Sachin Hambarde, Praful Shukla, Amit |
| author_facet | Setti, Ramakrishna Vamsi Rachapudi, Jagadeesh Chaudhary, Sachin Hambarde, Praful Shukla, Amit |
| contents | Large language models (LLMs) achieve impressive performance when a task is fully specified in a single turn, yet the same models lose up to 39% of that performance when the identical task is revealed incrementally across multiple turns, a phenomenon documented at scale as Lost in Conversation. Crucially, this collapse is almost entirely a reliability failure; the best case, the aptitude only falls 16%, while the unreliability more than doubles (+112%). We argue that the root cause is structural, a flat conversation history assigns equal implicit weight to every prior turn, giving the model no signal to distinguish a critical constraint from incidental dialog. We present SeDT Sentence-transformer Decision-Transformer, a training-free inference-time method that resolves this by importing return-to-go conditioning from offline reinforcement learning. SeDT annotates each conversation shard with a cumulative relevance score derived from three complementary semantic, lexical, and positional signals and presents the full annotated history to the model at the final turn, without weight changes, without training data, and without discarding context. Evaluated on the Lost-in-Conversation benchmark in three LLMs and three generation tasks, SeDT outperforms the sharded baseline in all nine model-task combinations, with gains up to +37.7% in mean performance P and simultaneous reductions in unreliability in seven of the nine combinations. In short, telling the model which past turns matter is sufficient to substantially recover the performance lost in conversation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_26788 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SeDT: Sentence-Transformer Decision-Transformer Conditioning for Multi-Turn Conversation Reliability Setti, Ramakrishna Vamsi Rachapudi, Jagadeesh Chaudhary, Sachin Hambarde, Praful Shukla, Amit Computation and Language Artificial Intelligence Large language models (LLMs) achieve impressive performance when a task is fully specified in a single turn, yet the same models lose up to 39% of that performance when the identical task is revealed incrementally across multiple turns, a phenomenon documented at scale as Lost in Conversation. Crucially, this collapse is almost entirely a reliability failure; the best case, the aptitude only falls 16%, while the unreliability more than doubles (+112%). We argue that the root cause is structural, a flat conversation history assigns equal implicit weight to every prior turn, giving the model no signal to distinguish a critical constraint from incidental dialog. We present SeDT Sentence-transformer Decision-Transformer, a training-free inference-time method that resolves this by importing return-to-go conditioning from offline reinforcement learning. SeDT annotates each conversation shard with a cumulative relevance score derived from three complementary semantic, lexical, and positional signals and presents the full annotated history to the model at the final turn, without weight changes, without training data, and without discarding context. Evaluated on the Lost-in-Conversation benchmark in three LLMs and three generation tasks, SeDT outperforms the sharded baseline in all nine model-task combinations, with gains up to +37.7% in mean performance P and simultaneous reductions in unreliability in seven of the nine combinations. In short, telling the model which past turns matter is sufficient to substantially recover the performance lost in conversation. |
| title | SeDT: Sentence-Transformer Decision-Transformer Conditioning for Multi-Turn Conversation Reliability |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2605.26788 |