LLMs fail causal discovery due to a kernel obstruction in observational learning, but interventional agents using frozen LLMs in Bayesian loops succeed without training on causal graph benchmarks.
arXiv preprint arXiv:2305.12766 , year =
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Manifold steering along activation geometry induces behavioral trajectories matching the natural manifold of outputs, while linear steering produces off-manifold unnatural behaviors.
Multimodal ICL lags text-only ICL in few-shot settings due to weak cross-modal reasoning alignment and unreliable task mapping transfer, with an inference-stage method proposed to strengthen transfer.
citing papers explorer
-
Why LLMs Fail at Causal Discovery and How Interventional Agents Escape
LLMs fail causal discovery due to a kernel obstruction in observational learning, but interventional agents using frozen LLMs in Bayesian loops succeed without training on causal graph benchmarks.
-
Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior
Manifold steering along activation geometry induces behavioral trajectories matching the natural manifold of outputs, while linear steering produces off-manifold unnatural behaviors.
-
Why Multimodal In-Context Learning Lags Behind? Unveiling the Inner Mechanisms and Bottlenecks
Multimodal ICL lags text-only ICL in few-shot settings due to weak cross-modal reasoning alignment and unreliable task mapping transfer, with an inference-stage method proposed to strengthen transfer.