Sliding-window transformers without positional encodings are Turing complete because the sliding window breaks permutation symmetry and suffices to simulate Post machines via a constant-size histogram state.
hub
Token assorted: Mixing latent and text tokens for improved language model reasoning
18 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
18
Pith papers citing it
1
external citations · external index
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
L-layer transformers under Log-ICoT curriculum provably learn k-parity with poly(n) samples and log k stages, matching explicit CoT efficiency without inference overhead.
PearlVLA achieves SOTA on LIBERO by separating VLM representations into visual grounding and an iterative latent plan branch refined via world model queries and RefineNet with process-reward RL.
Dynamic Latent Routing jointly learns discrete latent codes, routing policies, and model parameters via dynamic search to match or exceed supervised fine-tuning by 6.6 points on average in low-data settings across four datasets and six models.
Post-Reasoning boosts LLM accuracy by reversing the usual answer-after-reasoning order, delivering mean relative gains of 17.37% across 117 model-benchmark pairs with zero extra cost.
Entropy-guided supertokens from BPE on reasoning traces compress LLM outputs by 8.1% on average across models and math benchmarks with no accuracy loss while exposing strategy differences between correct and incorrect traces.
DyCon dynamically controls reasoning depth in LRMs by modeling evolving difficulty from step-level embeddings, reducing redundant steps across multiple benchmarks.
ALAR trains LLM agents to perform most reasoning in a latent space supervised by actions and escalates to explicit CoT only when needed, cutting tokens by up to 84.6% while preserving accuracy on search and tool-use benchmarks.
SpecFlow represents intermediate visual thoughts in fixed-size DCT space and uses classifier-free guidance to steer updates from textual thoughts, achieving up to 2.1x lower computation and KV cache costs.
GLR formulates latent reasoning as geometric path approximation in pretrained embedding space and reports shorter LLM generations on math tasks without an explicit length penalty.
OneVL achieves superior accuracy to explicit chain-of-thought reasoning at answer-only latency by supervising latent tokens with a visual world model decoder that predicts future frames.
SeLaR selectively applies latent soft reasoning in LLMs via entropy gating and contrastive regularization, outperforming standard CoT on five benchmarks without training.
Survey mapping RL techniques onto LLM training and highlighting gaps in value-based, off-policy, and bootstrapping methods.
TARPO is a pure RL framework using a token-wise action router to switch between discrete token generation and latent reasoning in LLMs, with joint optimization showing outperformance on benchmarks.
A 22M-parameter hyperbolic model answers structured EHR questions with accuracy close to LLM-based systems (EHRXQA 89.5%, MIMIC-Instr 76.0%).
LEPO applies RL to continuous latent representations in LLMs by injecting Gumbel-Softmax stochasticity for diverse trajectory sampling and unified gradient estimation, outperforming existing discrete and latent RL methods.
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.
Latent-CURE introduces latent-space chain-of-thought reasoning and dual-asymmetric optimization to produce transparent, robust breast cancer diagnoses in imbalanced cohorts.
citing papers explorer
-
Rethinking the Role of Positional Encoding: Sliding-Window Transformers without PE Remain Turing Complete
Sliding-window transformers without positional encodings are Turing complete because the sliding window breaks permutation symmetry and suffices to simulate Post machines via a constant-size histogram state.
-
Transformers Provably Learn to Internalize Chain-of-Thought
L-layer transformers under Log-ICoT curriculum provably learn k-parity with poly(n) samples and log k stages, matching explicit CoT efficiency without inference overhead.
-
PearlVLA: Progressive Embodied Action-Plan Refinement in Latent Space
PearlVLA achieves SOTA on LIBERO by separating VLM representations into visual grounding and an iterative latent plan branch refined via world model queries and RefineNet with process-reward RL.
-
Dynamic Latent Routing
Dynamic Latent Routing jointly learns discrete latent codes, routing policies, and model parameters via dynamic search to match or exceed supervised fine-tuning by 6.6 points on average in low-data settings across four datasets and six models.
-
Post Reasoning: Improving the Performance of Non-Thinking Models at No Cost
Post-Reasoning boosts LLM accuracy by reversing the usual answer-after-reasoning order, delivering mean relative gains of 17.37% across 117 model-benchmark pairs with zero extra cost.
-
Shorthand for Thought: Compressing LLM Reasoning via Entropy-Guided Supertokens
Entropy-guided supertokens from BPE on reasoning traces compress LLM outputs by 8.1% on average across models and math benchmarks with no accuracy loss while exposing strategy differences between correct and incorrect traces.
-
DyCon: Dynamic Reasoning Control via Evolving Difficulty Modeling
DyCon dynamically controls reasoning depth in LRMs by modeling evolving difficulty from step-level embeddings, reducing redundant steps across multiple benchmarks.
-
Adaptive Latent Agentic Reasoning
ALAR trains LLM agents to perform most reasoning in a latent space supervised by actions and escalates to explicit CoT only when needed, cutting tokens by up to 84.6% while preserving accuracy on search and tool-use benchmarks.
-
Spectral-Progressive Thought Flow for Lightweight Multimodal Reasoning
SpecFlow represents intermediate visual thoughts in fixed-size DCT space and uses classifier-free guidance to steer updates from textual thoughts, achieving up to 2.1x lower computation and KV cache costs.
-
Geometric Latent Reasoning Induces Shorter Generations in LLMs
GLR formulates latent reasoning as geometric path approximation in pretrained embedding space and reports shorter LLM generations on math tasks without an explicit length penalty.
-
Xiaomi OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation
OneVL achieves superior accuracy to explicit chain-of-thought reasoning at answer-only latency by supervising latent tokens with a visual world model decoder that predicts future frames.
-
SeLaR: Selective Latent Reasoning in Large Language Models
SeLaR selectively applies latent soft reasoning in LLMs via entropy gating and contrastive regularization, outperforming standard CoT on five benchmarks without training.
-
Modularized Reinforcement Learning on LLMs: From MDP Creation to Exploration and Learning
Survey mapping RL techniques onto LLM training and highlighting gaps in value-based, off-policy, and bootstrapping methods.
-
TARPO: Token-Wise Latent-Explicit Reasoning via Action-Routing Policy Optimization
TARPO is a pure RL framework using a token-wise action router to switch between discrete token generation and latent reasoning in LLMs, with joint optimization showing outperformance on benchmarks.
-
HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering
A 22M-parameter hyperbolic model answers structured EHR questions with accuracy close to LLM-based systems (EHRXQA 89.5%, MIMIC-Instr 76.0%).
-
LEPO: Latent Reasoning Policy Optimization for Large Language Models
LEPO applies RL to continuous latent representations in LLMs by injecting Gumbel-Softmax stochasticity for diverse trajectory sampling and unified gradient estimation, outperforming existing discrete and latent RL methods.
-
Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.
-
Latent-CURE for Breast Cancer Diagnosis
Latent-CURE introduces latent-space chain-of-thought reasoning and dual-asymmetric optimization to produce transparent, robust breast cancer diagnoses in imbalanced cohorts.