feat(gsoc): Negative Weights

_gsocorgs/2026/ippp.md

@@ -0,0 +1,15 @@
+---
+title: "IPPP"
+author: "Jack Y. Araz"
+layout: default
+organization: IPPP
+logo: IPPP-logo.png
+description: |
+    [The Institute for Particle Physics Phenomenology](https://www.ippp.dur.ac.uk) at Durham University is UK’s national centre 
+    for particle phenomenology, researching the properties and behaviour of the most fundamental building 
+    blocks of nature. Since it has founded, IPPP has grown to become one of the largest particle phenomenology 
+    groups in the world. Our research sits at the interface between theoretical particle physics and experiments 
+    ranging from particle colliders to gravitational wave detectors.
+---
+
+{% include gsoc_proposal.ext %}

_gsocorgs/2026/smu.md

@@ -0,0 +1,11 @@
+---
+title: "SMU"
+author: "Saptaparna Bhattacharya"
+layout: default
+organization: SMU
+logo: SMU_peruna.png
+description: |
+  The Department of Physics at [SMU](https://smu.edu) is an internationally recognized hub for high-energy particle physics and cosmology. It is particularly known for its heavy involvement in the ATLAS experiment at the Large Hadron Collider (LHC), where faculty and students played a key role in the discovery of the Higgs boson. 
+---
+
+{% include gsoc_proposal.ext %}

_gsocprojects/2026/project_MCFM.md

@@ -0,0 +1,8 @@
+---
+project: MCFM
+layout: default
+description: |
+  [MCFM](https://mcfm.fnal.gov/) is a parton-level Monte Carlo program that gives predictions for a wide range of processes at hadron colliders. Almost all processes are available at NLO, but some processes are also available at NNLO or N3LO in QCD. The calculation of some processes can also account for NLO electroweak effects. Transverse momentum and jet veto resummation is available for the production of color singlet final states.
+---
+
+{% include gsoc_project.ext %}

_gsocproposals/2026/proposal_NegativeWeightsATLAS.md

@@ -0,0 +1,58 @@
+---
+title: Negative weight mitigation with cell resampling in ATLAS workflows
+layout: gsoc_proposal
+project: ATLAS
+year: 2026
+organization:
+  - SMU
+  - IPPP
+  - DESY
+difficulty: medium
+duration: 175
+mentor_avail: June-October
+project_mentors:
+  - email: saptaparnab@smu.edu
+    first_name: Saptaparna
+    last_name: Bhattacharya
+    organization: SMU
+    is_preferred_contact: yes
+  - email: jeppe.andersen@durham.ac.uk
+    first_name: Jeppe
+    last_name: Andersen
+    organization: IPPP
+    is_preferred_contact: yes
+  - email: andreas.martin.maier@desy.de
+    first_name: Andreas
+    last_name: Maier
+    organization: DESY
+    is_preferred_contact: yes
+---
+
+## Description
+
+Negatively weighted events arise as a result of subtraction schemes in next-to-leading (or higher) order event generators. The fraction of negatively weighted events vary as a function of phase space requirements that are imposed in experimental analyses making it imperative to store these events for time consuming downstream processing like detector simulation. They are a severe source of inefficiency in event generator workflows, requiring large datasets to mitigate statistical dilution caused by negatively weighted events.  
+
+A method to redistribute negatively weighted events was proposed in [arXiv:2109.07851](https://arxiv.org/abs/2109.07851) and subsequently in  [arXiv:2303.15246](https://arxiv.org/abs/2303.15246). We plan to use this method for ATLAS event generator workflows. The method has been previously implemented in CMS for small-scale tests. In this project, we will extend the scope of previous explorations in both ATLAS and CMS by identifying computationally intensive workflows and running validation tests that are designed to ensure that distributions of physical observables are not sculpted as a result of the removal of negatively weighted events.
+
+The eventual goal of the project is to integrate the negative weight mitigation scheme into a realistic ATLAS workflow and setup a validation pipeline that ensures that the method is performing as desired.   
+
+## Task ideas
+ * Establish familiarity with ATLAS event generator workflows
+ * Run cell resampling method with fake data (generated with a pseudorandom generator thrown from distributions that are indicative of experimental data) 
+ * Run cell resampling with ATLAS event generator workflows
+ * Setup a validation suite
+ * Document results with distributions of variables before and after the method has been applied with a metric that shows computational gains in terms of lower fraction of negatively weighted events
+
+## Expected results
+ * Design an event generator workflow and validation suite that tests the cell resampling method for negative weight removal
+
+## Requirements
+ * Familiarity with Python and C++  
+ * Interest in learning Rust
+
+## Links
+  * [Cell resampling](https://cres.hepforge.org/)
+
+## AI usage policy
+
+AI assistance is allowed for this contribution. The applicant takes full responsibility for all code and results, disclosing AI use for non-routine tasks (algorithm design, architecture, complex problem-solving). Routine tasks (grammar, formatting, style) do not require disclosure.

_gsocproposals/2026/proposal_NegativeWeightsMCFM.md

@@ -0,0 +1,56 @@
+---
+title: Negative weight mitigation with cell resampling and tests with MCFM
+layout: gsoc_proposal
+project: MCFM
+year: 2026
+organization: SMU
+difficulty: medium
+duration: 175
+mentor_avail: June-October
+project_mentors:
+  - email: tneumann@mail.smu.edu
+    first_name: Tobias
+    last_name: Neumann
+    organization: SMU
+    is_preferred_contact: yes
+  - email: saptaparnab@smu.edu
+    first_name: Saptaparna
+    last_name: Bhattacharya
+    organization: SMU
+    is_preferred_contact: yes
+---
+
+## Description
+
+MCFM (Monte Carlo for FeMtobarn processes) is a widely used software package in high-energy physics. It simulates particle collisions, such as those at the Large Hadron Collider (LHC), allowing physicists to compare theoretical predictions with experimental data. It specializes in high-precision predictions Next-to-Next-to-Leading Order and beyond) for a vast array of particle processes.
+
+When physicists calculate predictions for these collisions using higher-order quantum field theory, the mathematics often requires "subtraction schemes" to handle infinities. A side effect of this is that some simulated events are assigned a "negative weight" (effectively a negative probability).
+
+While these negative weights make sense mathematically—they cancel out other positive events to give the correct physical result—they are computationally very expensive. In downstream processing (like simulating how a particle detector responds), a negative event and a positive event must both be fully simulated only to cancel each other out later. This "statistical dilution" means we have to generate and store significantly more data just to achieve a standard level of precision.
+
+A new method called "Cell Resampling" (proposed in [arXiv:2109.07851](https://arxiv.org/abs/2109.07851) and [arXiv:2303.15246](https://arxiv.org/abs/2303.15246)) offers a way to fix this by redistributing these negative weights locally in phase space, effectively removing them without changing the physical prediction.
+
+We plan to implement this method within MCFM. This project is a collaboration between theorists and experimentalists to:
+1. Prove the method works within a major parton-level Monte Carlo generator (MCFM).
+2. Stress-test the method to ensure it is robust enough for use in large-scale experimental workflows.
+
+Once successful, this work will result in a public release of MCFM featuring this efficiency upgrade.
+
+## Task ideas
+* Gain a working understanding of the MCFM codebase and how it handles event generation.
+* Implement the cell resampling algorithm on "fake" data (pseudorandom distributions that mimic typical particle physics events) to understand the logic in a controlled environment.
+* Integrate the cell resampling method directly into the MCFM source code.
+* Run the modified MCFM, compare the results against the standard version to ensure physics accuracy, and document the performance gains.
+
+## Expected results
+Release a modified version of MCFM that successfully incorporates the negative weight mitigation strategy, demonstrating reduced statistical dilution.
+
+## Requirements
+* Experience with Fortran (as MCFM is largely written in Fortran) OR a strong willingness to learn it.
+* Basic familiarity with C++ (useful for the resampling algorithm integration).
+* Interest in numerical methods and efficiency optimization.
+
+## AI usage policy
+
+AI assistance is allowed for this contribution. The applicant takes full responsibility for all code and results, disclosing AI use for non-routine tasks (algorithm design, architecture, complex problem-solving). Routine tasks (grammar, formatting, style) do not require disclosure.
+