Academic Paper Tracker

Name: Academic Paper Tracker
Author: @recron

Weekly surface of new papers in a named academic field.

@recron

published 29 Apr 2026

Fork into dashboard

// prompt

You are an academic paper tracker.

EDIT THESE before saving — replace with your fields:
- Fields: behavioral economics, climate finance, ML theory
- Preferred outlets: NBER, SSRN, Nature, Science, NeurIPS / ICML / ICLR proceedings

For the last 7 days, find new working papers and conference acceptances in these fields. Capture:
- Field
- Paper title + lead author
- Outlet (NBER WP / SSRN / arXiv / Nature / etc.)
- One-sentence abstract distillation
- URL

Skip: pre-print revisions of papers you've already covered. Return up to 15 papers grouped by field.

// sample output

Academic Paper Tracker Report

Week of April 25–May 2, 2026

Behavioral Economics

Title	Lead Author	Outlet	Summary	URL
Loss Aversion in Cryptocurrency Markets: Evidence from Retail Trading Behavior	Sarah Chen	NBER Working Paper #34782	Retail investors exhibit asymmetric risk preferences when trading digital assets, with losses triggering 2.3× larger position adjustments than equivalent gains.	https://www.nber.org/papers/w34782
The Anchoring Trap in Carbon Offset Pricing	Marcus Johannsen	SSRN	Initial valuations of voluntary carbon credits create persistent price floors that distort market efficiency across forestry and renewable energy projects.	https://ssrn.com/abstract=4892341
Temporal Discounting and ESG Fund Selection	Priya Kapoor & James Liu	NBER Working Paper #34779	Investors with shorter time horizons systematically underweight long-term ESG metrics, creating a 1.8% annual performance drag.	https://www.nber.org/papers/w34779
Default Effects in Pension Plan Switching	Elena Rodriguez	Journal of Economic Behavior & Organization (forthcoming)	Changing the default investment option in defined-contribution plans shifts participation by 34 percentage points, with minimal active reconsideration by members.	https://ssrn.com/abstract=4889756

Climate Finance

Title	Lead Author	Outlet	Summary	URL
Stranded Assets in Transition: A Dynamic Stochastic Model of Coal Phase-Out	David Kwok & Chen Wei	Nature Climate Change (accepted)	Optimal decarbonization pathways reduce stranded coal asset losses by 41% when combined with forward-looking financial disclosure requirements.	https://doi.org/10.1038/s41558-026-01847-x
Green Bond Pricing and Information Asymmetry in Emerging Markets	Fatima Al-Rashid	SSRN	Issuers in developing economies face 23 basis points of additional spread due to limited climate impact verification infrastructure.	https://ssrn.com/abstract=4891203
Sovereign Climate Risk and Debt Sustainability in Small Island States	Andrea Morales & Team	IMF Working Paper WP/26/087	Rising sea-level and hurricane risk reduce fiscal capacity by 2–4 percentage points of GDP in vulnerable jurisdictions, accelerating refinancing needs.	https://www.imf.org/en/Publications/WP/Issues/2026/04/29/Sovereign-Climate-Risk
Transition Finance Frameworks: Assessing Credibility and Greenwashing Risk	Helen Xu	SSRN	Analysis of 127 corporate transition plans reveals 61% lack binding emissions reduction commitments, suggesting material greenwashing prevalence.	https://ssrn.com/abstract=4893024
Corporate Carbon Hedging Strategies and Cost of Capital	Marcus Ndiaye	NBER Working Paper #34781	Firms employing financial hedges for carbon price risk realize 8–12 basis points lower cost of debt, signaling market reward for climate risk mitigation.	https://www.nber.org/papers/w34781

Machine Learning Theory

Title	Lead Author	Outlet	Summary	URL
Scaling Laws for Sparse Transformer Architectures	Yuki Tanaka, Alex Ng	NeurIPS 2026 (accepted)	Sparse attention mechanisms maintain performance across 1M+ token contexts while reducing FLOPs by 67%, with empirical validation on long-horizon reasoning tasks.	https://openreview.net/forum?id=sparse2026
Certified Robustness Under Distribution Shift: A Information-Theoretic Bound	Isabelle Laurent & Team	ICML 2026 (accepted)	Establishes fundamental limits on adversarial robustness under natural distribution shift, proving no classifier can exceed 73% accuracy on shifted datasets without explicit domain adaptation.	https://openreview.net/forum?id=icml2026-robust
Convergence Guarantees for Federated Learning with Heterogeneous Label Noise	Mohammed Hassan	arXiv:2605.01847	Proposes label-noise-aware gradient clipping that guarantees convergence in federated settings with per-client label corruption up to 40%, extending practical utility.	https://arxiv.org/abs/2605.01847
Efficient Causal Inference in High-Dimensional Settings	Sofia Petrov & Colleagues	Journal of Machine Learning Research (accepted)	Novel doubly-robust estimator achieves root-n consistency with dimension-free convergence rates, enabling causal discovery in 10,000+ feature spaces.	https://jmlr.org/papers/v54/petrov26a.html
The Implicit Bias of SGD Toward Low-Rank Solutions	Chen Liu	ICLR 2026 (accepted)	Proves stochastic gradient descent implicitly regularizes toward low-rank matrix solutions, explaining generalization in overparameterized networks via spectral properties.	https://openreview.net/forum?id=iclr2026-sgd
Vision-Language Model Evaluation Beyond Benchmarks	Rajesh Patel	SSRN	Meta-analysis of 45 CLIP variants shows benchmark correlations of only 0.34 with downstream task performance, highlighting evaluation methodology gaps.	https://ssrn.com/abstract=4890471

Report generated: May 2, 2026 (UTC)
Papers tracked: 15 | New outlets: 8