REVIEW 10 cited by
Homoclinic Floer homology via direct limits
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
Assume $M$ to be $\mathbb R^2$ or a closed surface of genus $g \geq 1$ and $\omega$ a symplectic form on $M$. Let $\varphi: M \to M$ be a symplectomorphism with hyperbolic fixed point $x$ and transversely intersecting stable and unstable manifolds $W^s(\varphi, x)$ and $ W^u(\varphi, x)$. The intersection points $W^s(\varphi, x) \cap\ W^u(\varphi, x)=:{\mathcal H}(\varphi, x)$ are called homoclinic points, and the (un)stable manifolds of a symplectomorphism are Lagrangian submanifolds. For this Lagrangian intersection problem with its wildly oscillating Lagrangian manifolds and infinite number of intersection points, we introduced in earlier works Floer homologies generated by so-called (semi)primary homoclinic points and analysed their dynamical and geometric properties. In this paper, we significantly generalise these earlier results by first defining a Floer homology generated by finite sets of contractible homoclinic points. These Floer homologies nevertheless still consider rather `local' aspects of $W^s(\varphi, x) \cap\ W^u(\varphi, x)$ since their generator sets are finite (but the number of contractible homoclinic points is infinite). To overcome this issue, we construct a direct limit of these `local' homoclinic Floer homologies over suitable index sets. These direct limits accumulate the information gathered by the finitely generated `local' homoclinic Floer homologies.
Forward citations
Cited by 10 Pith papers
-
DanceChat: Large Language Model-Guided Music-to-Dance Generation
An LLM-generated text choreography, fused with music and beat features, guides a diffusion model to produce more diverse and physically plausible dance motion, with a multi-modal alignment loss intended to bridge musi...
-
E3D-Bench: A Benchmark for End-to-End 3D Geometric Foundation Models
E3D-Bench compares 16 3D geometric foundation models on depth, reconstruction, pose, and view-synthesis tasks with a unified evaluation toolkit.
-
MemGuide: Intent-Driven Memory Selection for Goal-Oriented Multi-Session LLM Agents
MemGuide retrieves and filters past dialogue memories by intent and missing slots, and on its new synthetic benchmark MS-TOD it improves task success by 11 points and shortens dialogues by 2.84 turns.
-
AtomBench: A Benchmarking Framework for Generative Crystal Reconstruction Models in Conventional Superconductors
On two superconductor datasets, CDVAE best reproduces lattice parameters while AtomGPT (full text) or MatterGen (abstract) best reproduces atomic coordinates, but the comparison is confounded by unequal input information.
-
SPARK: Search Personalization via Agent-Driven Retrieval and Knowledge-sharing
SPARK is a proposed framework that coordinates persona-based LLM agents with layered memory and three collaboration protocols to produce personalized search.
-
OpenReview Should be Protected and Leveraged as a Community Asset for Research in the Era of Large Language Models
The paper advocates protecting and leveraging OpenReview's peer review corpus as a community asset for LLM-based review assistance, benchmarks, and alignment.
-
Zero-Direction Probing: A Linear-Algebraic Framework for Deep Analysis of Large-Language-Model Drift
A claim that null-space leakage yields provable LLM drift detection is undermined by an inverted eigenvalue inference and a sketched regret proof, with no empirical validation.
-
Stability and Thermodynamics of a Generalized Power-Law Dark Energy Model
A generalized power-law dark energy model, combining a linear pressure term with a power-law correction, admits a stable phantom-like accelerating attractor and satisfies the generalized second law precisely when the ...
-
Dynamic Long Short-Term Memory Based Memory Storage For Long Horizon LLM Interaction
A lightweight preference-memory system for LLMs is proposed, but its LSTM memory encoder shows no improvement in preference following and only the BERT preference filter performs moderately on formal utterances.
-
Model Inversion Attacks on Llama 3: Extracting PII from Large Language Models
Simple prompts to Llama 3.2 1B produce text that resembles PII, but the paper does not establish that this text is memorized training data.
Discussion (0). Sign in to comment.