DB-3DME supplies a human-rated 3D mesh dataset and shows that fine-tuning the visual encoder of Qwen-2.5-VL-7B produces automatic evaluations that align better with humans than prior VLMs.
hub
arXiv preprint arXiv:2409.12183 , year=
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
hub tools
citation-role summary
background 4
citation-polarity summary
roles
background 4polarities
background 4representative citing papers
LoRi distills implicit chain-of-thought by matching low-rank structures in hidden states, raising math-reasoning accuracy toward explicit CoT levels on LLaMA and Qwen models.
CrystalReasoner combines LLM reasoning traces with physical priors and multi-objective RL to generate valid, stable, and property-conditioned crystal structures.
ReactBench benchmark shows MLLMs suffer over 30% performance drop on complex topological reasoning tasks versus basic ones when evaluated on chemical reaction diagrams.
HiRO-Nav adaptively triggers reasoning only on high-entropy actions via a hybrid training pipeline and shows better success-token trade-offs than always-reason or never-reason baselines on the CHORES-S benchmark.
ShadowCoT introduces a reasoning-level backdoor attack on LLMs achieving 94.4% attack success rate and 88.4% hijacking success rate with 0.15% parameter updates via internal state conditioning and reasoning chain pollution.
Early entropy dynamics during LLM decoding mark when explicit reasoning becomes beneficial, enabling the training-free EDRM router that selects strategies per instance and yields 41-55% token savings with accuracy gains across 15 benchmarks.
CLPD improves LLM distillation for reasoning by combining explicit data curriculum with progressive teacher scheduling of increasing capacity.
Overthinking in medical QA is linearly decodable at 71.6% accuracy yet fixed residual-stream steering yields no correction across 29 configurations, while enabling selective abstention with AUROC 0.610.
Targeted prompting and system interventions enable local LLMs such as Llama 3.1 70B to exploit 83% of tested Linux privilege escalation vulnerabilities.
Longer textual reasoning chains degrade MLLM accuracy on fine-grained visual tasks; a new normalization and constrained-reward training framework mitigates the effect and sets new SOTA numbers.
Proposes a three-layer framework using formal AI reasoning for verification, derivation, and discovery in wireless communications theory.
citing papers explorer
-
DB-3DME: From Dataset to Benchmark for Human-aligned Automatic 3D Mesh Evaluation
DB-3DME supplies a human-rated 3D mesh dataset and shows that fine-tuning the visual encoder of Qwen-2.5-VL-7B produces automatic evaluations that align better with humans than prior VLMs.
-
LoRi: Low-Rank Distillation for Implicit Reasoning
LoRi distills implicit chain-of-thought by matching low-rank structures in hidden states, raising math-reasoning accuracy toward explicit CoT levels on LLaMA and Qwen models.
-
CrystalReasoner: Reasoning and RL for Property-Conditioned Crystal Structure Generation
CrystalReasoner combines LLM reasoning traces with physical priors and multi-objective RL to generate valid, stable, and property-conditioned crystal structures.
-
ReactBench: A Benchmark for Topological Reasoning in MLLMs on Chemical Reaction Diagrams
ReactBench benchmark shows MLLMs suffer over 30% performance drop on complex topological reasoning tasks versus basic ones when evaluated on chemical reaction diagrams.
-
HiRO-Nav: Hybrid ReasOning Enables Efficient Embodied Navigation
HiRO-Nav adaptively triggers reasoning only on high-entropy actions via a hybrid training pipeline and shows better success-token trade-offs than always-reason or never-reason baselines on the CHORES-S benchmark.
-
ShadowCoT: Cognitive Hijacking for Stealthy Reasoning Backdoors in LLMs
ShadowCoT introduces a reasoning-level backdoor attack on LLMs achieving 94.4% attack success rate and 88.4% hijacking success rate with 0.15% parameter updates via internal state conditioning and reasoning chain pollution.
-
When Do LLMs Reason? A Dynamical Systems View via Entropy Phase Transitions
Early entropy dynamics during LLM decoding mark when explicit reasoning becomes beneficial, enabling the training-free EDRM router that selects strategies per instance and yields 41-55% token savings with accuracy gains across 15 benchmarks.
-
Curriculum Learning-Guided Progressive Distillation in Large Language Models
CLPD improves LLM distillation for reasoning by combining explicit data curriculum with progressive teacher scheduling of increasing capacity.
-
Decodable but Not Corrected by Fixed Residual-Stream Linear Steering: Evidence from Medical LLM Failure Regimes
Overthinking in medical QA is linearly decodable at 71.6% accuracy yet fixed residual-stream steering yields no correction across 29 configurations, while enabling selective abstention with AUROC 0.610.
-
Enhancing Linux Privilege Escalation Attack Capabilities of Local LLM Agents
Targeted prompting and system interventions enable local LLMs such as Llama 3.1 70B to exploit 83% of tested Linux privilege escalation vulnerabilities.
-
Can Textual Reasoning Improve the Performance of MLLMs on Fine-grained Visual Classification?
Longer textual reasoning chains degrade MLLM accuracy on fine-grained visual tasks; a new normalization and constrained-reward training framework mitigates the effect and sets new SOTA numbers.
-
Rethinking Wireless Communications through Formal Mathematical AI Reasoning
Proposes a three-layer framework using formal AI reasoning for verification, derivation, and discovery in wireless communications theory.
- ScheMatiQ: From Research Question to Structured Data through Interactive Schema Discovery