Papers with latent reasoning
DReCa: A General Task Augmentation Strategy for Few-Shot Natural Language Inference (2021.naacl-main)
Copied to clipboard
| Challenge: | Meta-learning has not yet succeeded in NLP due to the lack of a well-defined task distribution . meta-learners tend to overfit their adaptation mechanism and datasets are heterogeneous . |
| Approach: | They propose a method for decomposing datasets into Reasoning Categories to form additional high quality tasks. |
| Outcome: | The proposed method improves the accuracy of meta-learners by 1.5-4% across four few-shot NLI problems. |
Unveiling Internal Reasoning Modes in LLMs: A Deep Dive into Latent Reasoning vs. Factual Shortcuts with Attribute Rate Ratio (2025.emnlp-main)
Copied to clipboard
Yiran Yang, Haifeng Sun, Jingyu Wang, Qi Qi, Zirui Zhuang, Huazheng Wang, Pengfei Ren, Jing Wang, Jianxin Liao
| Challenge: | Existing research in multi-hop questions has identified two reasoning modes, but has not investigated how these modes differ during inference. |
| Approach: | They propose a classification metric that compares latent reasoning and factual shortcuts in multi-hop questions. |
| Outcome: | The proposed metric achieves 90% accuracy on the proposed datasets and demonstrates effectiveness in RAG conflict scenarios. |
NeuroSym-Cal: Bridging the Reasoning-Execution Gap in Code Generation via Hierarchical Calibration (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing calibration methods rely on the assumption that consensus implies correctness . Existing methods fail under systematic errors, leading to miscalibrated high-confidence predictions. |
| Approach: | They propose a hierarchical calibration framework that measures confidence at two levels . they propose sensitivity analysis to measure local curvature of deductive process . |
| Outcome: | The proposed framework de-saturates overconfident errors and improves selective generation performance on OOD benchmarks. |
LEPO: Latent Reasoning Policy Optimization for Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing latent reasoning methods that use chain of thought (CoT) are limited to selecting one discrete token at each reasoning step, which potentially induces information loss. |
| Approach: | They propose a framework that injects controllable stochasticity into latent reasoning via Gumbel-Softmax, restoring LLMs' exploratory capacity and enhancing their compatibility with Reinforcement Learning (RL). |
| Outcome: | The proposed framework preserves richer information for more comprehensive reasoning and is compatible with Reinforcement Learning (RL). |
Hopping Too Late: Exploring the Limitations of Large Language Models on Multi-Hop Queries (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) can solve complex multi-step problems, but little is known about how these computations are implemented internally. |
| Approach: | They propose a "back-patching" analysis method to solve multi-hop queries . they propose resolving the bridge entity into the bridge and the second hop into the target entity into latent steps. |
| Outcome: | The proposed method solves multi-hop queries that require two information extraction steps . it shows that the later layers lack the necessary knowledge to correctly generate the answer . |
AnchorSeg: Language Grounded Query Banks for Reasoning Segmentation (2026.acl-long)
Copied to clipboard
Rui Qian, Chuanhang Deng, Qiang Huang, Jian Xiong, Mingxuan Li, Yingbo Zhou, Wei Zhai, Jintao Chen, Dejing Dou
| Challenge: | Existing models rely on a single segmentation token whose hidden state implicitly encodes both semantic reasoning and spatial localization . Existing methods rely only on SEG>, which encodes semantic reasoning, limiting the model's ability to explicitly disentangle what to segment from where to segment. |
| Approach: | They propose a method which reformulates reasoning segmentation as a structured conditional generation process over image tokens conditioned on language grounded query banks. |
| Outcome: | The proposed model bridges token-level predictions and pixel-level supervision by decoupling spatial grounding from semantic reasoning through structured language grounded query banks. |
Efficient Prior-Guided Reasoning for Robust Retrieval-Augmented Generation under Conflicts (2026.acl-long)
Copied to clipboard
| Challenge: | Retrieval-Augmented Generation (RAG) has become a standard paradigm for grounding Large Language Models (LLMs) however, performance degrades substantially when faced with noisy, outdated, or conflicting retrieved information. |
| Approach: | They propose a framework that explicitly elicits the model’s parametric knowledge as prior information to guide reasoning on retrieved documents. |
| Outcome: | The proposed framework achieves robust performance across varying degrees of external inconsistency and noise. |
Large Reasoning Models Are (Not Yet) Multilingual Latent Reasoners (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent work shows that large reasoning models arrive at the correct answer before completing textual reasoning steps, indicating the presence of latent reasoning. |
| Approach: | They conduct a systematic investigation of multilingual latent reasoning in large reasoning models across 11 languages. |
| Outcome: | The proposed model arrive at the correct answer before completing the reasoning steps, indicating the presence of latent reasoning. |
SpiralThinker: Latent Reasoning through an Iterative Process with Text–Latent Interleaving (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing latent reasoning methods lack mechanisms to ensure stable reasoning dynamics in latent space and a systematic way to interleave implicit and explicit reasoning. |
| Approach: | They propose a framework that performs iterative updates over latent representations while enabling interleaved reasoning across latent and textual steps. |
| Outcome: | SpiralThinker achieves state-of-the-art among latent reasoning baselines. |
Parallel Test-Time Scaling for Latent Reasoning Models (2026.acl-long)
Copied to clipboard
| Challenge: | Parallel test-time scaling is a pivotal approach for enhancing large language models. |
| Approach: | They propose two uncertainty-inspired stochastic strategies for parallel test-time scaling for latent reasoning models and a Latent Reward Model for aggregation. |
| Outcome: | The proposed model scales well with compute and enables effective trajectory selection. |